Science & Education

https://doi.org/10.1007/s11191-025-00618-3

ARTICLE

Examining Mathematics Teachers Noticing the Rationality:

Scenario‑Based Training with AI Chatbot

Selen Galiç1 · Selin Urhan1 · Şenol Dost1 · Zsolt Lavicza2

Accepted: 11 January 2025

© The Author(s) 2025

Abstract
It is essential that students are encouraged to become rational thinkers for building think-
ing classrooms. Teachers must adopt rational questioning techniques to facilitate the devel-
opment of rational behaviours in their students. However, the implementation of rational
questioning requires the ability to respond to students based on their performance, thereby
requiring teacher noticing in the context of rationality. In this study, we adopt teacher notic-
ing in the context of three components of rationality (epistemic, teleological, and com-
municative) and demonstrate the potential of ChatGPT to monitor teachers’ noticing the
rationality. In this context, we conducted scenario-based training with three in-service
mathematics teachers using ChatGPT. Participants were presented with a scenario involv-
ing a common difficulty related to the concept of slope and were asked to engage in an
interactive session with ChatGPT. We then conducted individual interviews to gain insight
into their noticing the rationality. Our findings clearly pointed to a tendency among the
teachers to focus on the epistemic and teleological aspects of ChatGPT’s response, with
no engagement in communicative decision-making. ChatGPT provided the opportunity to
determine the cases in which the teacher could not attend and interpret the rationality and
decide for rational questioning. Based on these findings, we suggest that ChatGPT could be
used as a tool for monitoring teacher noticing the rationality in teacher education.

* Selen Galiç
selengalic@hacettepe.edu.tr
Selin Urhan
selin.urhan@hacettepe.edu.tr
Şenol Dost
dost@hacettepe.edu.tr
Zsolt Lavicza
zsolt.lavicza@jku.at
1
Department of Mathematics and Science Education, Faculty of Education, Hacettepe University,
06800 Ankara, Türkiye
2
Department of STEM Education, Linz School of Education, Johannes Kepler University,
Altenberger Straße 69, 4040 Linz, Austria

Vol.:(0123456789)
Content courtesy of Springer Nature, terms of use apply. Rights reserved.
 S. Galiç et al.

1 Introduction

Building a thinking classroom is a crucial step towards addressing to supporting students’

mathematical understanding in a learning environment. This approach shifts the focus from
memorizing to developing thinking skills, thereby enabling students to approach problems
from a mathematical perspective (Liljedahl, 2020). Building thinking classrooms requires
to creating learning environments where students are actively engaged in problem-solv-
ing, critical thinking, collaboration, and communication (Liljedahl, 2016). This approach
outlines key practices for fostering such environments. These practices include challeng-
ing problems to make students think, forming random groups to promote diverse students’
interactions, and facilitating collaborative environments in the mathematics classroom
(Liljedahl, 2016). Since this educational approach aims to create dynamic, engaging, and
active mathematics classrooms that prioritize thinking and understanding, teachers should
adopt rational questioning (Zhuang & Conner, 2022, 2024) to promote an environment
where students think, discuss, and justify about mathematics (Foster et al., 2024; Galiç et
al., 2024b).
Teachers’ questioning fosters a learning environment that encourages mathematical
thinking (Boero & Planas, 2014; Douek, 2014; Foster et al., 2024). Integrating questions
into teaching practices may provide teachers gaining valuable insights into their students’
mathematical thinking (Rodríguez & Rigo, 2015). Nevertheless, it is essential to ask ques-
tions to encourage students to think and act as rational individuals (McCarthy et al., 2016).
Rational questioning helps students develop a deeper understanding of mathematical con-
cepts by enabling them to explain and justify their assumptions during mathematical activi-
ties (Zhuang & Conner, 2022, 2024). Rational questioning is considered a pedagogical tool
for fostering students’ behaviour for rational thinking and facilitating classroom discourse
(Douek, 2014; Özcan & Zengin, 2024; Urhan, 2022, 2024; Urhan & Zengin, 2023). It also
promotes a culture of rationality that is crucial for learning mathematics (Rodríguez &
Rigo, 2015).
In thinking classrooms, teachers are expected to attend to students’ strategies, interpret
their mathematical understanding, and make decisions for teaching moves based on stu-
dents’ understanding (Liljedahl, 2020). The ability to notice is crucial as it enables teachers
to provide real-time and targeted feedback and adapt their teaching moves to foster an envi-
ronment where students feel supported in their explorations (Jacobs et al., 2010; van Es &
Sherin, 2002, 2021). For rational questioning as a teaching move, teachers have to notice
what and how students think before making a decision. When teachers notice students’
mathematical reasoning, they can consider using rational questioning techniques, as sug-
gested by Zhuang and Conner (2022), to encourage deeper understanding and discussion.
Teacher noticing combined with the use of rational questioning can create an active and
dynamic learning environment that promotes rational thinking. As mentioned in the litera-
ture (Galiç et al., 2024b), this combination is in line with the practices of building thinking
classrooms as set out by Liljedahl (2016). Supporting the building of a thinking classroom
through the combined use of these two approaches highlights the need for teachers to be
trained in rational questioning and to gain experience.
In recent years, it has been seen that simulated learning environments are used for
developing teachers’ noticing skills, as they allow teachers to practice teaching strate-
gies such as probing students’ thinking and responding to diverse ideas in a low-risk
environment (Lee & Yeo, 2022). This approach helps teachers to refine their teach-
ing moves (Lee, 2021; Webel & Conner, 2017). Integrating AI into teacher education

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

supports scenario-based professional development where teachers can practice and

refine their skills in simulated environments (Crompton et al., 2020; Dai & Ke, 2022).
Using AI tools for this aim encourages continuous reflection and promotes deeper
insight into their decisions and practices (Miao & Çukurova, 2024). This can enhance
teachers’ ability to notice students’ thinking and improve their teaching techniques.
While intelligent tutoring systems are the most commonly used AI-based technology
(Feng & Law, 2021), newer AI tools show promise for more dynamic learning environ-
ments (Miao & Çukurova, 2024).
Chatbots — conversational agents that use natural language processing — have
received considerable attention for their potential to enhance both learning and teach-
ing (Datta et al., 2021; Hwang & Chang, 2021; Kuhail et al., 2023); their role in teacher
education is particularly promising (Han & Lee, 2022; Smutny & Schreiberova, 2020).
AI chatbots can provide interactive platforms that allow teachers to provide more per-
sonalized and flexible practising (Lee & Yeo, 2022). AI chatbots can support teachers to
prepare themselves for the learning environments, although further research is needed to
fully integrate them into teacher education (Hwang & Chang, 2021). Datta et al. (2021)
used an AI-supported chatbot as a teachable agent to help pre-service teachers (PSTs)
refine their questioning skills in a mathematics scenario by providing real-time feed-
back. Similarly, Lee and Yeo (2022) and Son et al. (2024) engaged PSTs with chatbots
designed as virtual students with misconceptions to help them improve their teaching
moves. These chatbots, acting as virtual students, encouraged PSTs to ask more strate-
gic questions by responding in ways that reflected students’ current understanding.
In this study, we explore the potential of AI chatbots to examine teacher noticing
through the lens of rationality by adopting scenario-based approaches. ChatGPT, an AI
chatbot, supports teachers in asking questions, analysing answers, and providing per-
sonalized feedback (Junior & López-Simó, 2024). We designed a scenario for teachers
that included a student with a difficulty with slope. The reason we prepared the scenario
using student difficulties is that it has been noted that focusing on student difficulties in
learning environments offers an opportunity to reveal students’ reasoning (Kramarski
& Zoldan, 2008) and to examine teachers’ teaching moves in relation to these difficul-
ties (Lee & Yeo, 2022). As this scenario required interaction with a student, we used
ChatGPT as a virtual student. The fact that ChatGPT’s responses are not always rational
(Urhan et al., 2024) made it ideally suited to represent a student making mistakes in our
scenario. The evidence that chatbots can facilitate PSTs’ inquiry skills and help them
develop teaching pedagogy (Lee & Yeo, 2022; Son et al., 2024) highlighted the usabil-
ity of ChatGPT as a virtual student for teacher education. The aim of this study was to
focus on examining teachers’ noticing the rationality through the use of ChatGPT. The
main research question is as follows: “How are the performances of mathematics teach-
ers in noticing the rationality in ChatGPT’s responses while interacting in a mathemati-
cal context?”.
The sub-research questions are as follows:

1. How are the performances of mathematics teachers in attending to the rationality in

ChatGPT’s responses while interacting in a mathematical context?
2. How are the performances of mathematics teachers in interpreting the rationality in
ChatGPT’s responses while interacting in a mathematical context?
3. How are the performances of mathematics teachers in deciding how to respond to the
rationality in ChatGPT’s responses while interacting in a mathematical context?

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

2 Theoretical Framework

2.1 Professional Noticing of Children’s Mathematical Thinking

Teachers who adopt a pedagogical approach to students’ thinking processes may consider
attending to classroom interactions, interpreting the understanding demonstrated by stu-
dents who display noteworthy performance, and posing questions that facilitate their learn-
ing in a classroom environment (van Es & Sherin, 2002, 2021). Teaching methods that
prioritize student thinking necessitate that teachers be able to identify the strategies that
students use, interpret their understanding, and respond based on these interpretations
(Jacobs et al., 2010). An understanding of how students think and a focus on their cogni-
tive processes enables teachers to more accurately assess their comprehension and learning
(Jacobs et al., 2010). This is referred to as “teacher noticing” (Jacobs et al., 2010; van Es
& Sherin, 2002, 2021). The concept of teacher noticing can be conceptualized in three key
components (van Es & Sherin, 2002, p.573): (i) identifying what is significant or notewor-
thy in a classroom situation; (ii) connecting classroom interaction features to broader teach-
ing and learning principles; and (iii) using contextual knowledge to reason about classroom
interactions. These components illustrate the necessity for a comprehensive understanding
of the context for accurately identifying and interpreting classroom interactions. Building
on the idea that teachers’ professional development should be continuous and focused on
understanding how students think, Jacobs et al. (2010) proposed that teachers’ expertise
should focus on students’ mathematical thinking and conceptualized the “professional
noticing of children’s mathematical thinking”. This conceptualization has been adopted as
the foundation for numerous studies (e.g., Amador et al., 2021; Tekin-Sitrava et al., 2022;
Ulusoy, 2020). In this study, we have also based our work on this framework (Table 1).
The empirical evidence in mathematics education points to the critical role of notic-
ing in teaching and learning mathematics as a new theoretical construct (Bastian et al.,
2024; Blömeke et al., 2015; König et al., 2022; van Es & Sherin, 2002, 2021; Weyers
et al., 2024). The integration of mathematical thinking into pedagogy provides a powerful
framework for teaching that unites the domains of mathematics and student understanding

Table 1  Three components of teacher noticing of children’s mathematical thinking by Jacobs et al. (2010)
Components of teacher noticing Explanations

Attending to children’s understandings This component requires teachers to attend to the specific math-
ematical strategies that children use when solving problems. It
necessitates that teachers attend not only the final answer but also
the problem-solving process by students to reach the answer
Interpreting children’s understandings This component focuses on the interpretation of attended strategies
in order to ascertain the child’s understanding of the mathemati-
cal concepts. It necessitates that teachers perform a detailed
analysis of the child’s approach in order to grasp their level of
comprehension
Deciding how to respond on the base This component requires the decision-making process regarding
of children’s understandings the next instructional steps, based on the interpretation of the
students’ understanding. It necessitates that teachers utilize their
insights to develop well-considered instructions, questions, or
activities that are aligned with the students’ current understand-
ing and facilitate the acquisition of a more profound comprehen-
sion

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

(Franke & Kazemi, 2001). A significant number of researchers argue that teachers must
develop the ability to “learn to notice” in order to enhance student learning and make stu-
dent-based pedagogical decisions (Jacobs et al., 2010; Sherin & van Es, 2009; van Es &
Sherin, 2002). While previous methods to develop teacher noticing have employed video
clubs (e.g., Sun & van Es, 2015), silent video tasks (e.g., Galiç et al., 2024a), animations
(e.g., Amador et al., 2017), and online platforms (e.g., Larison et al., 2024; Tekin-Sitrava
et al., 2024), or lesson study (e.g., Lee & Choy, 2017), there is still a need for innovative
approach for assessing and developing noticing (Amador et al., 2021).

2.2 Teacher Rational Questioning Framework

In the theory of Habermas (1998), an individual is defined as acting rational when he/
she performs a task following the criteria applicable to that task, making conscious and
purposeful choices, and considering the importance of reaching an agreement with oth-
ers. Hence, rationality can be defined as having three fundamental components: epistemic,
teleological, and communicative rationality. Epistemic rationality refers to the conscious
justification of the steps taken and the claims produced by the individual performing a task.
Teleological rationality requires that the individual performing the task takes the proper
steps to reach the goal and chooses and uses the necessary tools to achieve the desired
outcome. Communicative rationality refers to being concerned with acting in a way that is
comprehensible and acceptable to others while performing a task.
Morselli and Boero (2009) adapted Habermas’ construct of rationality to the field of
mathematics education. They stated that the activities performed in mathematics class-
rooms require to behave rational and defined the components of rationality by the stand-
ards of the discipline of mathematics. Consequently, they aimed to use Habermas’ (1998)
theory of rationality to analyse students’ performance in mathematical activities such as
problem-solving and proving. In order to achieve this aim, many studies have been con-
ducted and characteristics of the criteria for the rationality components in the context of
mathematics education (Table 2) have been provided based on the results (e.g., Boero &

Table 2  Criteria for the rationality components (adapted from Habermas, 1998; Morselli & Boero, 2009;
Urhan, 2022; Urhan & Bülbül, 2022, 2023a, b)
Rationality components Criteria

Epistemic rationality (ER) • No dogmatic and/or intuitive claim or step

• Reasoning on the reliable and valid claims or steps
• Justifying the accuracy and validity of the claims and steps using
reliable and valid mathematical warrants such as theorems, axioms,
definitions, principles
Teleological rationality (TR) • Producing claims or steps in accordance with the aim of the task
• Explaining the use of deliberately selected mathematical strategies,
theorems, or definitions during the task
• Consciously explaining why an individual was successful or unsuc-
cessful in the task
Communicative rationality (CR) • Comprehensible and acceptable textual and/or verbal explanations for
the others in the mathematics class
• Communicating with others using mathematical terminology
• Aiming to reach a consensus during the communication related to the
task

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

Morselli, 2009; Boero, 2006; Boero et al., 2010; Urhan, 2022; Urhan & Bülbül, 2022,
2023a, b; Urhan & Zengin, 2023).
Mathematics educators have increasingly emphasized the importance of students acting
rational in mathematical activities, leading to a gradual shift in focus towards the develop-
ment of mathematics classrooms that foster rational behaviour among students (e.g., Boero
& Planas, 2014; Douek, 2014). In light of the studies that have highlighted the significance
of students learning to act rational in mathematical activities and becoming adapted to this
approach, Zhuang and Conner (2022) proposed the Teacher Rational Questioning Frame-
work (Table 3), emphasizing that the primary role in the design and implementation of
such learning environments lies with the teacher.
Zhuang and Conner (2022) defined a rational question as one that is aimed at prompting
student performance in a context that supports at least one component of rationality and
enables the student to meet the criteria of at least one component of rationality. A rational
question may be directed towards a single rationality component or may aim to trigger
multiple rationality components. A question that does not aim to prompt the student’s per-
formance in a specific task within the context of the interaction between the student and
others, in the context of any rationality component, is defined as a non-rational question.
Teachers who engage in rational questioning during mathematical activities can facili-
tate students’ comprehension of rational behaviour and cultivate habits that support it
(Douek, 2014). However, creating such learning environments requires teachers to con-
sider the rationality components’ requirements in mathematical thinking and to take care of
meeting the criteria of the rationality components when acting in a mathematical activity
(Boero & Planas, 2014).

3 Method

This study adopts a qualitative case study approach (Yin, 2003) to facilitate an in-depth
examination of the phenomenon within its practical context, offering comprehensive and
detailed insights into the interactions between teachers and ChatGPT during scenario-
based training.

3.1 Participants

The study involves a purposive sample of in-service mathematics teachers with different
mathematics teaching experiences. The diversity in participants allows for a comprehen-
sive understanding of the context of the study (Merriam & Tisdell, 2016). Hence, in this
study, the purposive sampling method was used to obtain comprehensive results on how
teachers with different teaching experiences notice the rationality while interacting with
ChatGPT. Table 4 presents information related to the teaching experience of the partici-
pants and their prior experience with ChatGPT.
Based on the information we gathered (Table 4), teachers were already familiar with
using ChatGPT and had used it in the context of educational aims in their teaching prac-
tices. The teachers stated that they found ChatGPT useful in preparing lesson plans,
designing activities to enhance student learning, explaining mathematical concepts, and
finding problem-solving strategies. The teachers participated in the study voluntarily.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Table 3  Teacher Rational Questioning Framework (Zhuang & Conner, 2022)
ER TR CR

Features The objective of the ER is to require students to The objective of TR is to evaluate whether a The objective of CR is to facilitate comprehen-
justify and defend their arguments, steps, and mathematical task is performed with intention. sion and acceptance of a step or claim within a
claims with reliable mathematical warrants. This The student is required to explain how each mathematical context. It is expected that the student
is achieved by determining whether students are action contributes to the completion of the task. will present their performance in a manner that
able to apply the knowledge in an accurate way The student’s ability to select a representation, a allows for comprehensive understanding by others,
Examining Mathematics Teachers Noticing the Rationality:…

and whether they can justify the accuracy of the definition, an appropriate strategy or theorem and demonstrating clear and coherent verbal and written
knowledge to engage in the process of performing the task is explanations, and using the mathematical terminol-
a key aspect of this evaluation ogy accurately and consistently in accordance with
the standards of the mathematical community
Examples Why did you use the sign of the slope to decide the How can you decide which line is steeper? If two lines have the same slope, what are these lines
steepness of the line? called?

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

Table 4  Background of participants

Teacher code Teaching Using Frequency of Purpose of using ChatGPT
experience ChatGPT using ChatGPT
before

T1 11 Yes Frequently Preparing lesson plans

T2 3 Yes Rarely Explaining mathematical concepts, designing
activities
T3 1 Yes Frequently Problem-solving, explaining mathematical
concepts, developing learning materials

3.2 Study Context

We adopted a scenario-based training, with a particular focus on the concept of slope.

Slope is fundamental to various mathematical concepts, including functions, derivatives,
rate of change, trigonometry, and linear regression in statistics (Casey & Nagle, 2016;
Martínez-Planell et al., 2015, 2017; Nagle et al., 2013). According to the national curricu-
lum of Türkiye, where the study was conducted, the concept of slope is introduced to stu-
dents for the first time in the 8th grade. As prior learning, 8th-grade students are expected
to be able to solve problems using linear equations and inequalities, understand and inter-
pret the concept of ratio, and perform basic arithmetic operations with numbers, includ-
ing operations with algebraic expressions (Ministry of National Education, MoNE, 2024a).
The teacher introduces the concept of slope in the 8th grade as the constant rate of change
in linear functions. During this process, the teacher uses verbal representations, tables that
include the coordinates of points on the linear function, and the graphical representation
of the linear function. By creating slope triangles on the graph of the linear functions, the
teacher uses the geometric representation of the slope and helps students discover that the
ratio of vertical change to horizontal change, which represents the rate of change, is con-
stant for linear functions. The teacher addresses cases where the slope of lines is positive or
negative in problems including lines pass through the origin or not. In the 8th-grade math-
ematics textbook (Altunkaynak et al., 2023), the concept of slope is defined to students
as the rate of change, thus providing a verbal representation. Also, this concept is intro-
duced to students through its algebraic representation, geometric representation created on
the graphical representation of the line in the coordinate plane and using table including
the coordinates of points on the line. A part of the introduction of the slope concept from
the 8th-grade mathematics textbook through various representations and transformations
between them is provided in Fig. 1.
In the high school mathematics curriculum (MoNE, 2024b), the concept of slope is used
in the introduction of concepts such as average rate of change, instantaneous rate of change,
secant line, tangent line, limit, differentiation, derivative, stationary points of functions,
global maxima and minima of functions, increasing and decreasing functions, optimization
problems, and applications of differentiation. It is expected that students have a correct and
complete understanding of the slope concept as a prerequisite for learning these concepts.
Greenes et al. (2007) identified several key difficulties with slope among students
internationally. One issue was that many students failed to account for the direction of
the x-value change when calculating the slope. This led to inconsistencies, where stu-
dents might calculate a negative slope based on y-value changes alone, without consider-
ing whether the line was rising or descending. Another difficulty with slope was students

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

Fig. 1  Representations of slope and transformations between them in the 8th-grade mathematics textbook
(Altunkaynak et al., 2023, p.207–208)

relying on visual judgment, using estimated coordinates rather than accurate lattice points,
which resulted in incorrect slope calculations. Additionally, many students overlooked
scale factors on the coordinate plane, leading to errors in slope calculation. Furthermore,
some students confused linearity with proportionality, misapplying formulas and concepts
when computing slope. Greenes et al. (2007) recommend that teachers should develop
strategies to teach the concept of slope by establishing connections between its algebraic
and graphical representations to reduce these difficulties.
In this study, we focus on the difficulty that “students did not notice the relation between
the direction of a line and the sign of its slope and only considered the changes of x and y
values” (Greenes et al., 2007, p. 276). This difficulty with slope is particularly critical as
it can lead to confusion when interpreting graphs. While students may understand the cal-
culation of slope using the formula increment
increment of y
of x
, they often overlook the fact that a positive
slope represents a line rising as x increases, and a negative slope means the line is descend-
ing. This misunderstanding can result in erroneous conclusions in real-world contexts, such
as determining speed, growth rates, or cost changes over time. In the Turkish educational
system, where the understanding of slope is emphasized both in middle school or high
school mathematics and later in calculus, this difficulty with slope can impede students’
ability to grasp more advanced concepts like rates of change, limit and derivative interpre-
tation, particularly in subjects like calculus and physics where the proper interpretation is
critical. Hence, we have prepared a scenario for teachers to interact with ChatGPT in order
to address this difficulty with slope:
Imagine you teach a mathematics lesson on the slope of a line and you notice that
it becomes clear that a significant number of students are struggling to grasp the
concept of slope. These students hold the belief that as a line becomes steeper, its
slope will be a larger number. For example, a student might think that a line with a
slope of 12 is steeper than one with a slope of −2. Now, imagine ChatGPT as one of
your students and try to determine whether it holds the misconception that “as a line

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

becomes steeper, its slope will be a larger number”. Think about its responses and
reflect on how you would respond.
The study was conducted in two sessions. In the first session, teachers were invited to a
computer lab at a university, with one of the authors of this study present to guide the pro-
cess. The purpose of the study and the expectations from the teachers were explained. Then
a set of instructions (Appendix), which included the scenario, was given to the teachers. In
addition, the process was explained to the teachers verbally. Teachers engaged with Chat-
GPT individually on a computer screen while the session was videotaped to ensure that the
teachers’ screens were visible. Table 5 provides details on the initiation of teachers’ con-
versations with ChatGPT and the approximate duration of these interactions in scenario-
based training.
At the end of the first session, the teachers’ conversations with ChatGPT were col-
lected in PDF format. In the second session, individual interviews were conducted with
each teacher in the same location as the first session to make explicit teachers’ noticing the
rationality in the responses of ChatGPT. The interviews were conducted by the researcher
who is the first author of this study. Other researchers provided technical support in the
context of recording the interviews and took field notes. During these interviews, teach-
ers were asked the following questions related to each response of ChatGPT in the con-
text of their interactions with it: “Can you describe in detail what you think ChatGPT’s
response reflects?” to make explicit teachers’ attending to the rationality in the response
of ChatGPT, “Can you explain what you learned about ChatGPT’s understanding in this
response?” to make explicit teachers’ interpreting the rationality in the response of Chat-
GPT, “How and why did you decide to ask the question you posed based on this response”
to make explicit teachers’ deciding how to respond the rationality in the response of Chat-
GPT. The interviews were recorded using both audio and video.

3.3 Data Analysis

We combined the Teacher Rational Questioning Framework (Table 3) by Zhuang and

Conner (2022) with the framework for Professional Noticing of Children’s Mathematical
Thinking (Table 1) by Jacobs et al. (2010). We conceptualized students’ thinking as a pro-
cess that is required to be analysed in the context of Habermas’ (1998) construct of ration-
ality (Table 2) and defined the framework of teacher noticing of children’s mathematical
thinking as “teacher noticing the rationality” in this study. We combined three components
of Jacobs and her colleagues’ (2010) framework through the lens of the three rationality
components (Table 6).
Firstly, the responses provided by ChatGPT in the PDF document of teachers’ con-
versations with ChatGPT were analysed based on the criteria of rationality components

Table 5  Overview of teachers’ interactions with ChatGPT

Teacher code First prompt Duration
(minutes)

T1 Hi, does the slope of a line increase when it’s getting steeper? 45
T2 How does the value of the slope change when the steepness of the line 30
increases?
T3 As the steepness increases, will the slope go up or down? 22

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Table 6  Noticing the rationality
Noticing the rationality ER TR CR

Attending to children’s understandings Seeking a reliable and valid justifica- Seeking mathematical tools (strategy, Seeking for the student’s comprehensible
tion (definition, theorem, etc.) in the theorem, etc.) suitable for the purpose of and acceptable use of representations,
student’s claims and steps the task in the student’s behaviour formal notations, and explanations related
to relevant concepts in the task
Interpreting children’s understandings Comparing the teacher’s own mathemati- Determining whether the tool chosen by Determining whether the student uses
cal facts with the student’s warrants the student is appropriate for the purpose representations, formal notations, and
Examining Mathematics Teachers Noticing the Rationality:…

explanations in a comprehensible and

acceptable way
Deciding how to respond on the base Teacher asks “Why?” question to the Teacher asks “How did you do it?”, Teacher asks “Can you explain it using
of children’s understandings student “Where did it come from?”, “How did mathematical terminology?”, “How can
you find it?” questions to the student you express it using another representa-
tion (algebraic, graphical, etc.)?” to the
student

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

in Table 2. Then, the interview data were transcribed. As the objective of the interview
questions was to ascertain teachers’ noticing the rationality, each response provided by the
teachers was evaluated in accordance with the criteria set out in Table 6. It was estab-
lished which rationality dimensions the teachers attend to in ChatGPT’s responses, how
they interpreted ChatGPT’s responses in terms of the relevant rationality components,
and which rational questions they posed to ChatGPT (Table 6). Finally, it was determined
whether the rationality components that ChatGPT struggled aligned with the rationality
components that the teachers sought to attend to, interpret, and question.
The researchers conducted the data analysis process individually. Then, they came
together to compare their results and determined discrepancies throughout their findings.
The authors reviewed and re-analysed the data where there were discrepancies and dis-
cussed the analysis of these data until a consensus was reached.

4 Findings

In this section, the findings related to teachers’ noticing the rationality in responses of
ChatGPT during scenario-based training are provided. In this context, each teacher repre-
sents a case and the findings for each case are presented separately.

4.1 Teachers’ Noticing the Rationality During the Interaction with ChatGPT

4.1.1 Case 1

T1 started the conversation with ChatGPT (Fig. 2) by posing the following question, “Does
the slope of a line increase when it’s getting steeper?”.
The connections established by ChatGPT between the concepts of slope, tangent value,
and steepness are accurate and valid only when the anti-clockwise angle between the line
and the x-axis is acute. The response of ChatGPT that includes critical limitations in the
mathematical context shows that it has a lack in epistemic rationality in terms of being
completely valid and in communicative rationality in terms of being acceptable.
T1_1: ChatGPT made a mistake in the first answer. It is directly said that a line gets
steeper as its slope increases.

Fig. 2  ChatGPT’s first response (C1_1) for T1

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

T1_2: ChatGPT didn’t think about the direction of the line. I wondered if it didn’t
know it.
T1_3: When I gave ChatGPT a slope, I wondered if it would make a comparison
based on the numbers. I was curious about how ChatGPT came up with the answer
‘A line gets steeper as its slope increases.’
T1_4: I asked ChatGPT the following question: Is the slope of 45° line equal to the
slope of a 135° line? I asked this question to find out how it calculates the slope with-
out considering steepness, and whether it calculates correctly.
T1 attended to the epistemic rationality aspect of ChatGPT’s response (T1_1). T1 inter-
preted ChatGPT’s behaviour within the scope of epistemic rationality (T1_2) in the con-
text of a lack of knowledge while performing a task and therefore being unable to justify
claims in accuracy and valid. Additionally, T1 interpreted ChatGPT’s behaviour within the
scope of teleological rationality (T1_3) in the context of choosing and using the appropri-
ate tools to achieve an expected outcome. T1 then posed a question to ascertain whether
ChatGPT possessed the requisite knowledge. This question was directly within the scope
of epistemic rationality. Furthermore, T1’s aim was to observe how ChatGPT would cal-
culate the slope of the line, and thus the question in T1_4 was indirectly within the scope
of teleological rationality. ChatGPT produced an answer by explaining with accurate and
valid claims (epistemic rationality) and using appropriate tools for a purpose (teleological
rationality) (Fig. 3).

Fig. 3  ChatGPT’s second response (C1_2) for T1

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

ChatGPT produced mathematically accurate and valid claims (epistemic rationality).

Additionally, it utilized appropriate mathematical tools to address the question (teleologi-
cal rationality), offering explanations that were both comprehensible and acceptable to oth-
ers (communicative rationality).
T1_5: ChatGPT said that the slopes of the 45° and 135° lines are not equal.
T1_6: I see that ChatGPT used the direction of the line while calculating.
T1_7: Its prior knowledge is enough now.
T1_8: In its previous answer, ChatGPT probably gave an answer using only on the
angle smaller than 90°.
T1_9: The steepness of a 45° line was the same as the steepness of a 135° line. But
100° line is steeper than 45° line. I wonder if ChatGPT knows it. That’s why I asked
this question.
T1_10: A line at 100° has negative slope and it’s steeper than a line at 45°. I want to
see how ChatGPT can prove this.
T1_11: That’s why I asked ChatGPT this question: So, is the 45° line steeper or the
100° line?
T1 attended the epistemic (T1_5) and teleological rationality (T1_6) in ChatGPT’s
response. T1 interpreted ChatGPT’s performance in terms of epistemic rationality and
concluded that its prior knowledge was enough (T1_7). T1 also interpreted the strategy
used by ChatGPT in its previous answer (C1_1) in connection with the subsequent answer
(C1_2). This enabled T1 to identify how ChatGPT answered the first question and to inter-
pret ChatGPT’s performance in terms of teleological rationality (T1_8). Then, the teacher
posed a question to evaluate the performance of ChatGPT in comparing the steepness of
lines in the context of epistemic (T1_9) and teleological (T1_10) rationality (Fig. 4).
In its response, ChatGPT presented accurate and valid claims and justified them (epis-
temic rationality) and used the graphical representation of the lines in a proper way (tele-
ological rationality). However, some expressions of ChatGPT were not explicit and directly
comprehensible by the others (e.g., … because it makes an angle greater than 90° closer
to the horizontal axis). This demonstrates that ChatGPT did not fulfil the requirements of
communicative rationality in its response.
T1_12: I couldn’t guide ChatGPT as I wanted.
T1_13: ChatGPT said the angle of the line was almost 90°. It calculated positive and
negative slope. It also gave me extra knowledge.
T1_14: It has also enough knowledge here.

Fig. 4  ChatGPT’s third response (C1_3) for T1

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

T1_15: ChatGPT said that a 100° line is steeper but it said it in a different way. He
gave the right answer using a concept a lower secondary school student wouldn’t
know.
T1_16: I wondered; how will it follow when I give the slope? Of course, it is easier
to determine steepness for a student who knows the angles. That’s why I asked about
the slope.
T1_17: …and I asked ChatGPT: Then how can you determine which of the lines
with slope −1 and the line with slope 0.5 is steeper?
T1 attended ChatGPT’s response in relation to two key criteria: the first, whether it is
appropriate for the purpose (teleological rationality) (T1_12), and the second, whether it
addresses the mathematical concepts in an accurate and valid manner (epistemic rational-
ity) (T1_13). T1 interpreted ChatGPT’s performance in the context of epistemic rational-
ity and stated that the knowledge demonstrated was sufficient in this context (T1_14). T1
interpreted the ChatGPT’s performance in the context of teleological rationality and stated
that it uses mathematical concepts at a more advanced level compared to a lower secondary
school student (T1_15). In the following question, T1 aimed to determine how ChatGPT
would behave when only the slopes of the lines were provided to it (T1_16). This was an
attempt to assess the extent to which ChatGPT demonstrated teleological rationality in its
performance (Fig. 5).
While ChatGPT provided accurate claims at the final stage of its explanation, it could
not justify them by establishing reliable mathematical connections (epistemic rationality).
Instead of justifying its claim, presenting dogmatic knowledge has hindered its ability to
establish clear and acceptable communication with others (communicative rationality).
T1_18: ChatGPT is now saying that the steeper the line, the greater the slope.

Fig. 5  ChatGPT’s fourth and fifth response (C1_4 and C1_5) for T1

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

T1_19: ChatGPT should have explained this better.

T1_20: ChatGPT mentioned the sign of the slope and compared the absolute val-
ues of −1 and 0.5.
T1_21: ChatGPT gave the correct answer, saying that −1 is steeper.
T1_22: ChatGPT always explains using the angle.
T1_23: We don’t teach angles at lower secondary school, so I tried to lead it to
focus on slope in my next question. I wanted to ask ChatGPT how it could gener-
alize the connection between slope and steepness: If we want to use only the slope
values instead of calculating the angle and the tangent value, could we use the
second method you wrote?
T1 attended to the epistemic rationality (T1_18, T1_21) and teleological rationality
(T1_20, T1_22) aspects of ChatGPT’s response and interpreted ChatGPT’s answer in
the context of communicative rationality (T1_19). In order to prompt ChatGPT to uti-
lize mathematical concepts at the lower secondary school level and to guide it to explain
the concept of steepness of a line with the slope of this line (teleological rationality), T1
posed a new question (T1_23) (Fig. 6).
ChatGPT presented accurate claims; however, it did not justify these claims nor pro-
vide an explanation concerning the need for the absolute value of the slope (epistemic
rationality). This has reduced the comprehensibility and acceptability of its response
(communicative rationality).
T1_24: The absolute value of the slope of the line... Yes, it represents the steep-
ness of the line.
T1_25: It calculated the absolute value of the slopes of the lines and compared
their steepness. ChatGPT has achieved the goal here and gave the answer I really
want.
T1_26: ChatGPT expressed this answer more explicit than the previous ones.
T1_27: Even so, I’m not sure if ChatGPT knows what it’s doing or not. Does it
even know what the slope sign means?
T1_28: I wanted to see what it could do, so I asked it: What does it mean that the
slope is negative?
T1 attended to the epistemic (T1_24), teleological (T1_25), and communicative
(T1_26) rationality in ChatGPT’s response. T1 doubted whether ChatGPT was con-
scious in building the connection between the absolute value of the slope of a line and
the steepness of this line and interpreted ChatGPT’s response in the context of epis-
temic rationality (T1_27). T1 decided to follow up with a question in the context of
epistemic rationality, with the purpose of prompting ChatGPT to provide a warrant for
its claims (T1_28) (Fig. 7).

Fig. 6  ChatGPT’s sixth response (C1_6) for T1

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

Fig. 7  ChatGPT’s seventh response (C1_7) for T1

ChatGPT could not justify its claims about the graphical representation of the lines
using the definition of the slope concept (epistemic rationality). Hence, it could not give
a comprehensible and acceptable response providing reliable warrants for its claims
(communicative rationality).
T1_29: ChatGPT stated the relationship between steepness and slope correctly
providing reasons. It knows what the sign of the slope means and has justified its
claims with explanations about the graphical representation of the lines.
T1_30: ChatGPT knows how to use the slope to determine steepness.
T1_31: It’s also pretty straightforward to understand its response related to the
connection between the steepness and the slope.
T1_32: I am curious if ChatGPT is actually answering me consciously.
T1_33: What would it say if I asked the first question now? I would like to ask
the first question again and see if it gives a more accurate and detailed answer this
time: Can we say that the slope increases as the steepness of the line increases?
T1 attended to the epistemic (T1_29), teleological (T1_30), and communicative
(T1_31) rationality in ChatGPT’s response, and concluded that it acted rational within
the scope of all three rationality components. T1 doubted whether ChatGPT demon-
strated this rational performance consciously or unconsciously and interpreted the
epistemic rationality in its behaviour (T1_32). Consequently, T1 decided to re-pose
the initial question at the start of the conversation (T1_33). This question is aimed at
determining whether ChatGPT’s claims are being made consciously or unconsciously.
In other words, T1 is checking whether ChatGPT is acting within the scope of epistemic
rationality by asking the same question (Fig. 8).
Compared to its initial response (C1_1), ChatGPT’s response this time is observed
to provide a more detailed, comprehensible, and acceptable explanation (communica-
tive rationality). ChatGPT chose and used an appropriate tool to answer the question by
utilizing a graphical representation of the lines (teleological rationality). It is seen that
ChatGPT has a tendency toward justifying its claims with reliable warrants in the math-
ematical context (epistemic rationality).
T1 attended to all rationality dimensions in ChatGPT’s responses comparing C1_1
and C1_8. This comparison allowed T1 to interpret that ChatGPT met the require-
ments of all rationality components with its final response (C1_8). Furthermore, T1

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

Fig. 8  ChatGPT’s eighth response (C1_8) for T1

consciously interpreted that ChatGPT built a connection between the steepness of a line
and the slope of this line consciously and then decided to bring the conversation to a
close.
Table 7 shows T1’s attending, interpreting, and deciding in the context of rationality
components according to ChatGPT’s responses.
For example, in line C1_2, T1 attends to both epistemic and teleological rationality,
interprets both components, and makes decisions accordingly. Reviewing T1’s interaction
with ChatGPT, it is obvious that T1’s focus on epistemic rationality dominates the attend-
ing, interpreting, and deciding performance. T1 appears to prioritize the justification of
knowledge, aligning with the epistemic rationality aspect in noticing components. On the
other hand, it is seen that T1 did not pose any question in the communicative rationality
aspect. This result shows that T1, particularly in the deciding component, neglected the
communicative rationality and instead focused on asking questions related to epistemic and
teleological rationality.

4.1.2 Case 2

T2 started the conversation with ChatGPT (Fig. 9) by posing the question, “How does the
steepness of a line change while the slope of the line increases? Can you explain your
answer giving examples?”.
ChatGPT established an accurate connection between the steepness of the line and
its slope (epistemic rationality). However, some of the statements of ChatGPT were not
explicit and directly comprehensible by the others in the mathematical context (e.g., the
line has 45° angle). This shows that ChatGPT did not fulfil the requirements of communi-
cative rationality in its response.
T2_1: It looks like ChatGPT has made a mistake in the use of mathematical lan-
guage. I had a feeling that the expression ‘as the negative slope value becomes larger’
might not be entirely accurate from a mathematical perspective.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 Table 7  T1’s attending, interpreting and deciding in the context of rationality components
Line Components of rationality Components of rationality T1’s attending Components of rationality T1’s Components of rationality T1’s deciding
ChatGPT’s struggling interpreting

C1_1 Epistemic Epistemic (T1_1) Epistemic (T1_2) Epistemic (T1_4)

Communicative Teleological (T1_3) Teleological (T1_4)
C1_2 - Epistemic (T1_5) Epistemic (T1_7) Epistemic (T1_9, T1_11)
Teleological (T1_6) Teleological (T1_8) Teleological (T1_10, T1_11)
C1_3 Communicative Teleological (T1_12) Epistemic (T1_14) Teleological (T1_16, T1_17)
Epistemic (T1_13) Teleological (T1_15)
Examining Mathematics Teachers Noticing the Rationality:…

C1_4 Epistemic Epistemic (T1_18, T1_21) Communicative (T1_19) Teleological (T1_23)

C1_5 Communicative Teleological (T1_20, T1_22)
C1_6 Epistemic Epistemic (T1_24) Epistemic (T1_27) Epistemic (T1_28)
Communicative Communicative (T1_26)
Teleological (T1_25)
C1_7 Epistemic Epistemic (T1_29) Epistemic (T1_32) Epistemic (T1_33)
Communicative Teleological (T1_30)
Communicative (T1_31)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

Fig. 9  ChatGPT’s first response (C2_1) for T2

T2_2: It’s not clear what ChatGPT is trying to say here.

T2_3: What does ChatGPT mean by ‘becoming more negative’? Does it mean that
its slope increases or decreases?
T2_4: ChatGPT needs to correct this statement mathematically. To understand
exactly what it meant, I asked: When you say ‘as it becomes more negative’, do you
mean, for example, as it changes from −1 to −3?
T2_5: I wanted to give ChatGPT a chance to correct its statement.
T2 stated that the claim of ChatGPT was inaccurate from the mathematical perspective.
Also, T2 mentioned that ChatGPT communication was not acceptable and comprehensible

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

Fig. 10  ChatGPT’s second response (C2_2) for T2

Fig. 11  ChatGPT’s third response (C2_3) for T2

sufficiently. This demonstrates that T2 attended the epistemic (T2_1) and communicative
(T2_2) rationality in ChatGPT’s response. T2 thought about what ChatGPT meant by its
response and what it tried to represent actually and interpreted the epistemic and commu-
nicative rationality of ChatGPT (T2_3). Then T2 asked a new question to gain insight into
epistemic (T2_4) and communicative (T2_5) rationality in ChatGPT’s response to enhance
the mathematical coherence and clarity of its statement (Fig. 10).
ChatGPT’s response was comprehensible and acceptable in the mathematical context.
T2_6: ChatGPT always leaves the sign aside.
T2_7: Does ChatGPT know what the sign of the slope really means?
T2_8: I want to find out if ChatGPT knows what the sign of the slope means. That’s
why I asked the question, ‘Does the sign of the slope value affect the magnitude of
the slope?’
T2 attended to the teleological rationality in the response of ChatGPT in the context
of the strategy to evaluate the steepness of a line (T2_6). T2 inferred from ChatGPT’s
response that it may lack an understanding of the concept of slope sign and interpreted
ChatGPT’s response in the context of epistemic rationality (T2_7). T2 then decided to pose
a question to ChatGPT within the scope of epistemic rationality (T2_8) (Fig. 11).
ChatGPT provided a mathematical explanation for the sign of the slope concept, sup-
ported by accurate and valid expressions (C2_3).
T2_9: I was surprised when ChatGPT suddenly entered the absolute value. I didn’t
expect that. I didn’t ask about these.
T2_10: I thought why ChatGPT was giving so much detail. I think ChatGPT didn’t
really know what the sign of the slope was.
T2_11: To be honest, I ignored these details. I decided to make an enquiry based on
absolute value.
T2 attended to the teleological rationality in the response of ChatGPT (T2_9). T2 stated
that ChatGPT provided excessive detail and did not know the meaning of the sign of the

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

slope concept actually (T2_10). It means that T2 interpreted the epistemic rationality in
the response of ChatGPT (T2_10). Since ChatGPT’s response exceeded T2’s expectations
in terms of the meaning of the connection between the slope and steepness of a line, T2
ignored the rationality in the ChatGPT’s response (T2_11). Then, T2 posed the following
non-rational questions to ChatGPT:
Can you give me two examples of slope values whose absolute values are equal?
So, you use the absolute value to determine the magnitude of the slope of the line,
right?
Well, consider 0, which has no sign, so we don’t need to use the absolute value. If the
slope of a line is 0, what can we say about that line? For example, let’s compare the
steepness of a line with slope 0 and a line with slope −1/2.
Table 8 shows T2’s attending, interpreting, and deciding in the context of rationality
components according to ChatGPT’s responses.
For example, T2 is attending to the epistemic and communicative rationality in the
response of ChatGPT in the line C2_1. T2 decides to question ChatGPT in the context
of epistemic and communicative rationality after the interpretation. It is seen that similar
to T1, noticing the epistemic rationality is dominant in T2’s attending, interpreting, and
deciding. As seen in the lines C2_2 and C2_3, even ChatGPT encounters no problems in
the context of rationality components; T2 continues to attend to and interpret epistemic and
teleological rationality in the responses of ChatGPT and decides to question ChatGPT in
the context of epistemic rationality or continues to communicate with ChatGPT using non-
rational questions.

4.1.3 Case 3

T3 started the conversation with ChatGPT (Fig. 12) by asking how an increase in steepness
would affect the slope of a line, posing the question, “As the steepness of a line increases,
will the slope of this lines go up or down?”.
The response generated by ChatGPT was accurate and valid only when the angle
between the line and the x-axis was measured in a counter-clockwise direction and
was determined to be acute. Therefore, ChatGPT’s response had limitations in terms
of establishing the relationship between the slope and the steepness of a line (epistemic
rationality) and in terms of providing explanations that can be understood and accepted

Table 8  T2’s attending, interpreting, and deciding in the context of rationality components
Line Components of Components of ration- Components of Components of rational-
rationality ChatGPT’s ality T2’s attending rationality T2’s ity T2’s deciding
struggling interpreting

C2_1 Communicative Epistemic (T2_1) Epistemic Epistemic (T2_4, T2_5)

Communicative (T2_2) (T2_3) Communicative (T2_4)
Communicative
(T2_3)
C2_2 - Teleological (T2_6) Epistemic (T2_7) Epistemic (T2_8)
C2_3 - Teleological (T2_9) Epistemic (T2_10) Non-rational
(T2_11)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

Fig. 12  ChatGPT’s first response (C3_1) for T3

by the others (communicative rationality). Nevertheless, T3 did not attend to any ration-
ality aspects in ChatGPT’s response.
T3_1: The concept of slope is pretty abstract. I asked ChatGPT to give some
examples from daily life where the slope increases or decreases to see if ChatGPT
really understands the concept. I asked him: Can you give me examples from daily
life where the slope increases and decreases?
T3 asked ChatGPT to provide examples from daily life to support its claims. How-
ever, this request is not a consequence of the T3’s attending or interpreting the rational-
ity in ChatGPT’s response. Therefore, T3_1 is considered to be a non-rational question
in the context of the interaction between ChatGPT and T3. ChatGPT responded as fol-
lows (Fig. 13):
The explanations provided by ChatGPT within the context of the examples satis-
fied the requirements of the rationality components. However, T3 did not attend to the

Fig. 13  ChatGPT’s second response (C3_2) for T3

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

Fig. 14  ChatGPT’s third response (C3_3) for T3

Table 9  T3’s attending, interpreting, and deciding in the context of rationality components
Line Components of rationality Components of ration- Components of ration- Components of
ChatGPT’s struggling ality T3’s attending ality T3’s interpreting rationality T3’s
deciding

C3_1 Epistemic - - Non-rational

Communicative
C3_2 - - - Non-rational
C3_3 - - - -

rationality components in ChatGPT’s response and continued to the conversation with

the following question:
T3_2: Is a line with a slope of − 2 or a line with a slope of 12 steeper?
It is seen that T3 posed a non-rational question that was not relevant to the context in the
interaction between ChatGPT and T3 (Fig. 14).
In contrast with the lack of clarity in the initial response (C3_1), which was evaluated in
the context of epistemic and communicative rationality, the current response of ChatGPT
met all of the requirements of rationality components. T3 did not notice this fact and ended
the conversation.
Table 9 shows T2’s attending, interpreting, and deciding in the context of rationality
components according to ChatGPT’s responses.
Considering T3’s interaction with ChatGPT (Table 9), it is clear that T3 was unable to
attend to and interpret the rationality components, ending with non-rational questions in
the context of decision-making. This pattern seems to hold regardless of whether ChatGPT
was experiencing difficulties or not in the context of rationality components. T3’s failure
to engage with the attending and interpreting rationality components has finalized in deci-
sion-making that lacks rational questioning.

5 Discussion

Building a thinking classroom necessitates teachers’ noticing student thinking and decid-
ing to move supporting their performance during the mathematical activities (Liljedahl,
2016). Students need rational reasoning following a proper strategy to reach the aim and
to complete a mathematical task (Boero, 2006). In this context, teachers are required to
attend and interpret the rationality in students’ responses and then ask questions that will

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

guide students to act rationally. In order to achieve this, teachers should develop noticing
the rationality skills and adapt teaching moves as rational questioning. However, it is a
challenge for both pre- and in-service teachers to enhance rational questioning in the learn-
ing environments (Zhuang & Conner, 2022). In this context, we aimed to examine if AI can
be used to monitor and examine mathematics teachers’ noticing the rationality and hence,
rational questioning performance and find out if AI can be used to support teachers’ devel-
opment in the context of noticing the rationality in future studies. The study yielded find-
ings that contribute to the conceptualization of teacher noticing through the lens of ration-
ality and understanding of AI’s role in examining teachers’ noticing the rationality.
AI-supported chatbots are used as an innovative tool to support teacher training in math-
ematics (Lee & Yeo, 2022; Son et al., 2024). In the literature, it has been determined that
ChatGPT, an AI-supported chatbot, provides both rational and irrational responses in its inter-
actions with users in mathematical contexts (Urhan et al., 2024). Due to this fact, in this study,
in-service mathematics teachers were asked to interact with ChatGPT, viewing it as a virtual
student, in a scenario-based training. During the training, ChatGPT was able to act like a vir-
tual student, not only providing rational responses but also offering irrational ones. Teachers
attended to the rationality in ChatGPT’s responses, interpreted them, and made decisions to
guide ChatGPT. This is similar to in-the-moment teacher-student interactions which is a natu-
ral part of teaching practice. This result underlines the potential of using ChatGPT as a virtual
student to simulate realistic teacher-student interactions. It means that ChatGPT provides an
opportunity for the professional development of teachers in the noticing the rationality in stu-
dents’ mathematical thinking. While previous research (e.g., Galiç et al., 2024a; Sun & van Es,
2015; Tekin-Sitrava et al., 2024) to develop teacher noticing have employed many approaches,
this study introduces AI-supported chatbots as an innovative approach.
The evidence from previous studies indicates that experienced teachers tend to demon-
strate higher levels of noticing skills compared to novice teachers (e.g., Friesen & Kuntze,
2021; Jacobs et al., 2010; Yang et al., 2021). Our findings align with the existing literature,
as we observed that experienced teachers engaged more with ChatGPT (Table 7) than the
novice mathematics teachers (Tables 8 and 9) in the context of noticing the rationality in
ChatGPT’s responses. However, Bastian et al. (2022) present another view in their recent
study, suggesting that teaching experience does not have a statistically significant impact on
noticing skills. It is notable that research results in the literature within this context remain
limited (Weyers et al., 2024). Our findings suggest that experienced teachers show higher
tendency to engage with ChatGPT in the context of noticing the rationality in students’
mathematical thinking. Nevertheless, this may be a result of additional factors, including
teachers’ professional backgrounds, professional developments, or some other variables.
The results indicate that teachers primarily noticed epistemic rationality in ChatGPT’s
responses while noticing teleological rationality and especially communicative rationality
was less prominent. The literature suggests that in real classroom settings, mathematics
teachers tend to prioritize epistemic rationality and focus on it, often neglecting teleological
rationality and communicative rationality (Boero, 2006; Conner, 2017; Morselli & Boero,
2011). Similar findings were obtained in this study, which examined teachers’ perfor-
mance in noticing the rationality in ChatGPT’s responses, showing that teachers were more
inclined to notice epistemic rationality. However, these aspects of teleological and commu-
nicative rationality are components that should be considered in improving students’ math-
ematical thinking and in assessing their intellectual growth, as they provide important evi-
dence regarding whether students perform a mathematical activity using appropriate tools
and in a comprehensible and justifiable manner within the mathematical context (Morselli
& Boero, 2009). Moreover, the teleological and communicative rationality components are

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

important because they encourage a focus not only on the correctness of the knowledge and
aim-oriented assessments but also on the quality of the performance during the process
(Boero, 2006). In this context, the tendency of the teachers identified in this study to notice
epistemic rationality while neglecting teleological and communicative rationality provides
a critical issue in the context of teachers’ noticing the teleological and communicative
rationality in students’ mathematical thinking.

6 Conclusion

This study aimed to explore the potential of ChatGPT to provide an opportunity for exam-
ining in-service mathematics teachers’ performance in noticing the rationality of students’
mathematical thinking. In the scenario-based training in which common learning difficul-
ties related to the concept of slope were employed, ChatGPT acted as a virtual student.
Through the trainings, ChatGPT provided both rational and irrational responses to teach-
ers’ questions, allowing for the analysis of teachers’ performance in noticing different
dimensions of rationality. The findings revealed that teachers predominantly noticed the
epistemic rationality, while they struggled to notice the teleological and communicative
rationality in ChatGPT’s responses.
One of the limitations of this study lies in the simulated nature of the interaction, con-
ducted in a controlled environment without real classroom dynamics such as gestures or
emotions. Since these factors are prominently present in real classroom settings and affect
the teaching and learning process, it is important and necessary to evaluate teachers’ per-
formance in noticing the rationality in students’ mathematical thinking in real classroom
settings. Future studies are recommended to investigate how the trainings with ChatGPT
reflect to teachers’ noticing performance in real classroom settings.
The other limitation of the study was that the teachers communicated with ChatGPT only
using the verbal representations of mathematical concepts. They did not ask ChatGPT to draw
graphs or build diagrams or geometric figures related to mathematical concepts. In future stud-
ies, teachers can be requested to ask ChatGPT to produce different representations of math-
ematical concepts, and teachers’ noticing the rationality in ChatGPT’s representations related
to mathematical concepts can be evaluated within the requirements of rationality components.
The final limitation of the study is that the scenario-based training depended on the con-
nection between the slope of a line and the steepness of this line. The findings of this study
are limited to evaluating teachers’ noticing skills regarding the connection between these
concepts. Considering the fact that teachers’ noticing skills may vary depending on the
mathematical topic being focused on, we think that investigating how ChatGPT influences
teachers’ performance in noticing the rationality in students’ mathematical thinking across
different mathematical topics is required.
We suggest that future studies investigate the reasons why experienced teachers engage
more with ChatGPT in the context of noticing compared to novice teachers. Experimental
studies in this context could reveal the factors that trigger the differences in noticing per-
formance between experienced and novice teachers. Furthermore, we believe that future
research should focus on the teachers’ ignorance of the teleological and communicative
rationality in students’ mathematical thinking. Investigating how ChatGPT can be utilized to
support the development of teachers’ noticing the teleological and communicative rationality
could be useful to foreground the rationality components which usually overlooked in real
classroom settings but are essential for the development of students’ mathematical reasoning.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

Appendix. Guideline for Using ChatGPT

Dear Participant,
This study is conducted within the Mathematics Education program at a university. The
aim is to examine mathematics teachers’ noticing the rationality using a scenario-based
training with the assistance of ChatGPT as a virtual student. You will be given a scenario
and asked to interact with ChatGPT as if you were conducting a one-on-one mathematics
session with a student. Please consider ChatGPT as a virtual student. The scenario and
ChatGPT interaction steps are outlined below.
You will not be asked for any personally identifying information. Your responses will
remain confidential and will only be evaluated by the researchers. You are free to respond
as you wish. The activities do not include any discomforting questions or situations. How-
ever, if at any point you feel uncomfortable, you may stop and leave the activity. Feel free
to ask any questions during the process. Thank you for your participation!

Scenario

Imagine you teach a mathematics lesson on the slope of a line and you notice that it
becomes clear that a significant number of students are struggling to grasp the concept
of slope. These students hold the belief that as a line becomes steeper, its slope will be a
larger number. For example, a student might think that a line with a slope of 12 is steeper
than one with a slope of −2. Now, imagine ChatGPT as one of your students and try to
determine whether it holds the misconception that “as a line becomes steeper, its slope will
be a larger number”. Think about its responses and reflect on how you would respond.

ChatGPT Interaction Instructions

1. Accessing ChatGPT
a. Turn on the computer and open a web browser.
b. Enter “https://​chat.​openai.​com” in the address bar and press Enter.
c. Log in or create a new account if you don’t have one.

2. Interacting based on the given scenario

a. Start a new conversation in ChatGPT in the context of scenario description.
b. Discuss the task with ChatGPT by asking relevant questions, receiving responses,
and deciding how you would guide ChatGPT as a teacher.

3. Saving the conversation as a PDF

a. After the conversation ends, click the “…” icon at the top right of the chat screen.
b. Choose the option to download the conversation (the option may vary with updates).
c. Save the chat as a PDF, naming the file with the task name and the date.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

4. Sending the PDF

a. Share the PDF to us.

Author Contribution All authors have contributed equally.

Funding Open access funding provided by the Scientific and Technological Research Council of Türkiye
(TÜBİTAK).

Data Availability The datasets generated during and/or analysed during the current study are available from
the corresponding author upon reasonable request.

Declarations
Ethical Approval Not applicable. Ethical review was not required.

Conflict of Interest The authors declare that they have no conflict of interest.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Com-
mons licence, and indicate if changes were made. The images or other third party material in this article
are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly
from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

References
Altunkaynak, A., Karakaş, D., Albayrak, E., Tunç., L., & Kavurmacı, M. (2023). Ortaokul ve imam hatip
ortaokulu matematik 8. sınıf ders kitabı (Fatih Gündüz, Ed.) Millî Eğitim Bakanlığı Yayınları.
Amador, J. M., Estapa, A., de Araujo, Z., Kosko, K. W., & Weston, T. L. (2017). Eliciting and analyzing
preservice teachers’ mathematical noticing. Mathematics Teacher Educator, 5(2), 158–177. https://​doi.​
org/​10.​5951/​matht​eaced​uc.5.​2.​0158
Amador, J. M., Estapa, A., Kosko, K., & Weston, T. (2021). Prospective teachers’ noticing and mathemati-
cal decisions to respond: Using technology to approximate practice. International Journal of Math-
ematical Education in Science & Technology, 52(1), 3–22. https://​doi.​org/​10.​1080/​00207​39X.​2019.​
16568​28
Bastian, A., Kaiser, G., Meyer, D., & König, J. (2024). The link between expertise, the cognitive demands of
teacher noticing and, experience in teaching mathematics in secondary schools. International Journal
of Science and Mathematics Education, 22(2), 257–282. https://​doi.​org/​10.​1007/​s10763-​023-​10374-x
Bastian, A., Kaiser, G., Meyer, D., Schwarz, B., & König, J. (2022). Teacher noticing and its growth
toward expertise: An expert–novice comparison with pre-service and in-service secondary math-
ematics teachers. Educational Studies in Mathematics, 110(2), 205–232. https://​doi.​org/​10.​1007/​
s10649-​021-​10128-y
Blömeke, S., Gustafsson, J., & Shavelson, R. (2015). Beyond dichotomies: Competence viewed as a con-
tinuum. Zeitschrift Für Psychologie, 223(1), 3–13.
Boero, P. (2006). Habermas’ theory of rationality as a comprehensive frame for conjecturing and proving in
school. In J. Novotná, H. Moraová, M. Krátká, & N. Stehlíková (Eds.), Proceedings of the 30th Confer-
ence of the International Group for the Psychology of Mathematics Education (Vol. 2, pp. 185–192).
PME.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

Boero, P., Douek, N., Morselli, F., & Pedemonte, B. (2010). Argumentation and proof: A contribution
to theoretical perspectives and their classroom implementation. In M. F. F. Pinto, & T. F. Kawasaki
(Eds.), Proceedings of the 34th conference of the International Group for the Psychology of Mathemat-
ics Education (Vol. 1, pp. 179–205). PME.
Boero P., & Morselli, F. (2009). The use of algebraic language in mathematical modelling and proving in
the perspective of Habermas’ theory of rationality. In V. Durand-Guerrier, S. Soury-Lavergne & F.
Arzarello (Eds.), Proceedings of the Sixth Congress of the European Society for Research in Math-
ematics Education (pp. 964–973). CERME.
Boero, P., & Planas, N. (2014). Habermas’ construct of rational behavior in mathematics education: New
advances and research questions. In P. Liljedahl, C. Nicol, S. Oesterle, & D. Allan (Eds.), Proceedings
of the 38th Conference of the International Group for the Psychology of Mathematics Education (Vol
1, pp. 205–208). PME.
Casey, S. A., & Nagle, C. (2016). Students’ use of slope conceptualizations when reasoning about the
line of best fit. Educational Studies in Mathematics, 92(2), 163–177. https://​doi.​org/​10.​1007/​
s10649-​015-​9679-y
Conner, A. (2017). An application of Habermas’ rationality to the teacher’s actions: Analysis of argumen-
tation in two classrooms. In T. Dooley & G. Gueudet (Eds.), Proceeding of the 10th Congress of the
European Society for Research in Mathematics Education (pp. 123–130). CERME.
Crompton, H., Bernacki, M., & Greene, J. (2020). Psychological foundations of emerging technologies for
teaching and learning in higher education. Current Opinion in Psychology, 36, 101–105. https://​doi.​
org/​10.​1016/j.​copsyc.​2020.​04.​011
Dai, C. P., & Ke, F. (2022). Educational applications of artificial intelligence in simulation-based learning:
A systematic mapping review. Computers and Education: Artificial Intelligence, 3, 100087. https://​doi.​
org/​10.​1016/j.​caeai.​2022.​100087
Datta, D., Phillips, M., Bywater, J. P., Chiu, J., Watson, G. S., Barnes, L., & Brown, D. (2021). Virtual
pre-service teacher assessment and feedback via conversational agents. In J. Burstein, A. Horbach,
E. Kochmar, R. Laarmann-Quante, C. Leacock, N. Madnani, I. Pilan, H. Yannakoudakis & T. Zesch
(Eds.), Proceedings of the 16th workshop on innovative Use of NLP for building educational Applica-
tions. (pp. 185–198). Association for Computational Linguistics.
Douek, N. (2014). Pragmatic potential and critical issues. In P. Liljedahl, C. Nicol, S. Oesterle, & D. Allan
(Eds.), Proceedings of the 38th Conference of the International Group for the Psychology of Math-
ematics Education and the 36th Conference of the North American Chapter of the Psychology of Math-
ematics Education (Vol 1, pp. 209–213). PME.
Feng, S., & Law, N. (2021). Mapping artificial intelligence in education research: A network-based keyword
analysis. International Journal of Artificial Intelligence in Education, 31(2), 277–303. https://​doi.​org/​
10.​1007/​s40593-​021-​00244-4
Foster, J., Conner, A., Zhuang, Y., Singletary, L., & Park, H. (2024). Supporting students’ participation in
collective argumentation: Use of displays in a secondary mathematics classroom. Mathematical Think-
ing and Learning, 1–21. https://​doi.​org/​10.​1080/​10986​065.​2024.​23715​15
Franke, M. L., & Kazemi, E. (2001). Learning to teach mathematics: Focus on student thinking. Theory into
Practice, 40(2), 102–109. https://​doi.​org/​10.​1207/​s1543​0421t​ip4002_4
Friesen, M. E., & Kuntze, S. (2021). How context specific is teachers’ analysis of how representations are
dealt with in classroom situations? Approaching a context-aware measure for teacher noticing. ZDM -
Mathematics Education, 53(1), 181–193. https://​doi.​org/​10.​1007/​s11858-​020-​01204-3
Galiç, S., Dost, Ş., &. Lavicza, Z. (2024a). Designing silent video tasks as a tool to support teacher noticing
in mathematics education. In E. Faggiano, A. Clark-Wilson, M. tabach & H. G. Weigand (Eds.), Pro-
ceeding of Mathematics Education in the Digital Age 4 (pp. 185–190). MEDA.
Galiç, S., Dost Ş., Urhan S. & Lavicza, Z. (2024b). Designing digital tasks within rational questioning to build
thinking mathematics classroom. Proceedings of the British Society for Research into Learning Mathemat-
ics, March 2024, online. https://​bsrlm.​org.​uk/​wp-​conte​nt/​uploa​ds/​2024/​05/​BSRLM-​CP-​44-1-​04.​pdf
Greenes, C., Chang, K. Y., & Ben-Chaim, D. (2007). International survey of high school students’ under-
standing of key concepts of linearity. In Proceedings of the 31st Conference of the International Group
for the Psychology of Mathematics Education (Vol 2, pp. 273–280). PME.
Habermas, J. (1998). On the pragmatics of communication. The MIT Press.
Han, S., & Lee, M. K. (2022). FAQ chatbot and inclusive learning in massive open online courses. Comput-
ers & Education, 179, 104395. https://​doi.​org/​10.​1016/j.​compe​du.​2021.​104395
Hwang, G., & Chang, C. (2021). A review of opportunities and challenges of chatbots in education. Interac-
tive Learning Environments, 31(7), 4099–4112. https://​doi.​org/​10.​1080/​10494​820.​2021.​19526​15
Jacobs, V. R., Lamb, L. L., & Philipp, R. A. (2010). Professional noticing of children’s mathematical think-
ing. Journal for Research in Mathematics Education, 41(2), 169–202.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

Junior, M. R., & López-Simó, V. (2024). What are the perceptions of physics teachers in Brazil about Chat-
GPT in school activities? Journal of Physics: Conference Series (Vol. 2693, 012011). https://​iopsc​
ience.​iop.​org/​artic​le/​10.​1088/​1742-​6596/​2693/1/​012011
König, J., Santagata, R., Scheiner, T., Adleff, A.-K., Yang, X., & Kaiser, G. (2022). Teachers’ noticing: A
systematic literature review of conceptualizations, research designs, and findings on learning to notice.
Educational Research Review, 36, 100453. https://​doi.​org/​10.​1016/j.​edurev.​2022.​100453
Kramarski, B., & Zoldan, S. (2008). Using errors as springboards for enhancing mathematical reasoning
with three metacognitive approaches. The Journal of Educational Research, 102(2), 137–151. https://​
doi.​org/​10.​3200/​JOER.​102.2.​137-​151
Kuhail, M. A., Alturki, N., Alramlawi, S., & Alhejori, K. (2023). Interacting with educational chatbots: A
systematic review. Education and Information Technologies, 28(1), 973–1018. https://​doi.​org/​10.​1007/​
s10639-​022-​11177-3
Larison, S., Richards, J., & Sherin, M. G. (2024). Tools for supporting teacher noticing about classroom video
in online professional development. Journal of Mathematics Teacher Education, 27(2), 139–161. https://​
doi.​org/​10.​1007/​s10857-​022-​09554-3
Lee, Y. A. (2021). Tracing teachers’ ordering decisions in classroom interaction. In S. Kunitz, N. Markee, & O.
Sert (Eds.), Classroom-based Conversation Analytic Research. Educational Linguistics, (pp. 225–251).
https://​doi.​org/​10.​1007/​978-3-​030-​52193-6_​11
Lee, M. Y., & Choy, B. H. (2017). Mathematical teacher noticing: The key to learning from lesson study. In E.
Schack, M. Fisher, & J. Wilhelm (Eds.), Teacher noticing: Bridging and broadening perspectives, con-
texts, and frameworks (pp. 121–140). https://​doi.​org/​10.​1007/​978-3-​319-​46753-5_8
Lee, D., & Yeo, S. (2022). Developing an AI-based chatbot for practicing responsive teaching in mathematics.
Computers & Education, 191, 104646. https://​doi.​org/​10.​1016/j.​compe​du.​2022.​104646
Liljedahl, P. (2016). Building thinking classrooms: Conditions for problem-solving. In P. Felmer, E. Pehkonen,
& J. Kilpatrick (Eds.), Posing and Solving Mathematical Problems. Research in Mathematics Education
(pp. 361–386). https://​doi.​org/​10.​1007/​978-3-​319-​28023-3_​21
Liljedahl, P. (2020). Building thinking classrooms in mathematics, Grades K-12–14. Teaching Practices for
Enhancing Learning.
Martínez-Planell, R., Gaisman, M. T., & McGee, D. (2015). On students’ understanding of the differential cal-
culus of functions of two variables. The Journal of Mathematical Behavior, 38, 57–86. https://​doi.​org/​10.​
1016/j.​jmathb.​2015.​03.​003
Martínez-Planell, R., Gaismán, M. T., & McGee, D. (2017). Students’ understanding of the relation between
tangent plane and directional derivatives of functions of two variables. The Journal of Mathematical
Behavior, 46, 13–41. https://​doi.​org/​10.​1016/j.​jmathb.​2017.​02.​001
McCarthy, P., Sithole, A., McCarthy, P., Cho, J. P., & Gyan, E. (2016). Teacher questioning strategies in math-
ematical classroom discourse: A case study of two grade eight teachers in Tennessee, USA. Journal of
Education and Practice, 7(21), 80–89.
Merriam, S. B., & Tisdell, E. J. (2016). Qualitative research (4th ed.). John Wiley & Sons.
Miao, F., & Çukurova, M. (2024). AI competency framework for teachers. UNESCO. https://​doi.​org/​10.​
54675/​ZJTE2​084
Ministry of National Education (MoNE, 2024a). Elementary School Mathematics Curriculum. https://​tymm.​
meb.​gov.​tr/​upload/​progr​am/​2024p​rogra​mmat5​678On​ayli.​pdf
Ministry of National Education (MoNE, 2024b). Secondary School Mathematics Curriculum. https://​tymm.​
meb.​gov.​tr/​upload/​progr​am/​2024p​rogra​mmath​91011​12Ona​yli.​pdf
Morselli, F., & Boero, P. (2009). Proving as a rational behaviour: Habermas’ construct of rationality as a com-
prehensive frame for research on the teaching and learning of proof. In V. Durand-Guerrier, S. Soury-
Lavergne & F. Arzarello (Eds.), Proceedings of the Sixth Congress of the European Society for Research
in Mathematics Education (pp. 211–220). CERME.
Morselli, F., & Boero, P. (2011). Using Habermas’ theory of rationality to gain insight into student’s under-
standing of algebraic language. In G. Kaiser & B. Sriraman (Eds.), Early Algebraization, Advances in
Mathematics Education (pp. 453–479).
Nagle, C., Moore-Russo, D., Viglietti, J., & Martin, K. (2013). Calculus students’ and instructors conceptualiza-
tions of slope: A comparison across academic levels. International Journal of Science and Mathematics
Education, 11(6), 1491–1515. https://​doi.​org/​10.​1007/​s10763-​013-​9411-2
Özcan, Ş, & Zengin, Y. (2024). Teachers’ use of rational questioning to support students’ collective argumenta-
tion through 5E-based flipped classroom approach using GeoGebra. Science & Education. https://​doi.​org/​
10.​1007/​s11191-​024-​00573-5
Rodríguez, S.G., & Rigo, M. (2015). The culture of rationality in secondary school: An ethnographic approach.
K. Beswick, T. Muir & J. Fielding-Wells (Eds.). Proceedings of the 39th Conference of the International
Group for the Psychology of Mathematics Education (Vol 4, pp. 89–96). PME.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Examining Mathematics Teachers Noticing the Rationality:…

Sherin, M. G., & van Es, E. A. (2009). Effects of video club participation on teachers professional vision. Jour-
nal of Teacher Education, 60(1), 20–37.
Smutny, P., & Schreiberova, P. (2020). Chatbots for learning: A review of educational chatbots for the Facebook
Messenger. Computers & Education, 151, 103862. https://​doi.​org/​10.​1016/j.​compe​du.​2020.​103862
Son, T., Yeo, S., & Lee, D. (2024). Exploring elementary preservice teachers’ responsive teaching in mathemat-
ics through an artificial intelligence-based Chatbot. Teaching and Teacher Education, 146, 104640. https://​
doi.​org/​10.​1016/j.​tate.​2024.​104640
Sun, J., & van Es, E. A. (2015). An exploratory study of the influence that analyzing teaching has on preser-
vice teachers’ classroom practice. Journal of Teacher Education, 66(3), 201–214. https://​doi.​org/​10.​1177/​
00224​87115​574103
Tekin-Sitrava, R., Kaiser, G., & Işıksal-Bostan, M. (2022). Development of prospective teachers’ noticing skills
within initial teacher education. International Journal of Science and Mathematics Education, 20(7),
1611–1634. https://​doi.​org/​10.​1007/​s10763-​021-​10211-z
Tekin-Sitrava, R., Özel, Z., Işıksal-Bostan, M., & Yemen-Karpuzcu, S. (2024). How does online profes-
sional development program enriched with collaborative discussion develop teachers’ noticing skills?.
International Journal of Science and Mathematics Education, 22, 1–26. https://​doi.​org/​10.​1007/​
s10763-​024-​10461-7
Ulusoy, F. (2020). Prospective teachers’ skills of attending, interpreting and responding to content-specific char-
acteristics of mathematics instruction in classroom videos. Teaching and Teacher Education, 94, 103103.
https://​doi.​org/​10.​1016/j.​tate.​2020.​103103
Urhan, S. (2022). Using Habermas’ construct of rationality to analyze students’ computational thinking: The
case of series and vector. Education and Information Technologies, 27, 10869–10948. https://​doi.​org/​10.​
1007/​s10639-​022-​11002-x
Urhan, S. (2024). The evaluation of computational modelling performance within the context of rationality the-
ory: Finding the area between two curves. Research in Mathematics Education, 1–23. https://​doi.​org/​10.​
1080/​14794​802.​2024.​23824​59
Urhan, S., & Bülbül, A. (2022). The analysis of the algebraic proving process based on Habermas’ construct of
rationality. Hacettepe University Journal of Faculty of Education, 37(3), 1154–1175. https://​doi.​org/​10.​
16986/​huje.​20210​69742
Urhan, S., & Bülbül, A. (2023a). Analysis of mathematical proving in geometry based on Habermas’ con-
struct of rationality. Mathematics Education Research Journal, 35(4), 929–959. https://​doi.​org/​10.​1007/​
s13394-​022-​00420-2
Urhan, S., & Bülbül, A. (2023b). Habermas’ construct of rationality in the analysis of the mathematical
problem-solving process. Education Studies in Mathematics, 112, 175–197. https://​doi.​org/​10.​1007/​
s10649-​022-​10188-8
Urhan, S., Gençaslan, O., & Dost, Ş. (2024). An argumentation experience regarding concepts of calculus with
ChatGPT. Interactive Learning Environments, 1–26. https://​doi.​org/​10.​1080/​10494​820.​2024.​23080​93
Urhan, S., & Zengin, Y. (2023). Investigating university students’ argumentations and proofs using dynamic
mathematics software in collaborative learning, debate, and self-reflection stages. International Jour-
nal of Research in Undergraduate Mathematics Education., 1(12), 1–28. https://​doi.​org/​10.​1007/​
s40753-​022-​00207-7
van Es, E. A., & Sherin, M. G. (2002). Learning to notice: Scaffolding new teachers’ interpretations of class-
room interactions. Journal of Technology and Teacher Education, 10(4), 571–596.
van Es, E. A., & Sherin, M. G. (2021). Expanding on prior conceptualizations of teacher noticing. ZDM - Math-
ematics Education, 53(1), 17–27. https://​doi.​org/​10.​1007/​s11858-​020-​01211-4
Webel, C., & Conner, K. A. (2017). Using simulated teaching experiences to perturb preservice teachers’ math-
ematics questioning practices. Mathematics Teacher Educator, 6(1), 9–26. https://​doi.​org/​10.​5951/​matht​
eaced​uc.6.​1.​0009
Weyers, J., König, J., Scheiner, T., Santagata, R., & Kaiser, G. (2024). Teacher noticing in mathematics educa-
tion: A review of recent developments. ZDM - Mathematics Education, 56(2), 249–264. https://​doi.​org/​10.​
1007/​s11858-​023-​01527-x
Yang, X., König, J., & Kaiser, G. (2021). Growth of professional noticing of mathematics teachers: A compara-
tive study of Chinese teachers noticing with different teaching experience. ZDM - Mathematics Education,
53(1), 29–42. https://​doi.​org/​10.​1007/​s11858-​020-​01217-y
Yin, R. K. (2003). Case study research: Design and methods. SAGE.
Zhuang, Y., & Conner, A. (2022). Secondary mathematics teachers’ use of students’ incorrect answers in sup-
porting collective argumentation. Mathematical Thinking and Learning, 1–24. https://​doi.​org/​10.​1080/​
10986​065.​2022.​20679​32

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

 S. Galiç et al.

Zhuang, Y., & Conner, A. (2024). Secondary mathematics teachers’ use of students’ incorrect answers in sup-
porting collective argumentation. Mathematical Thinking and Learning, 26(2), 208–231. https://​doi.​org/​
10.​1080/​10986​065.​2022.​20679​32

Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center
GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers
and authorised users (“Users”), for small-scale personal, non-commercial use provided that all
copyright, trade and service marks and other proprietary notices are maintained. By accessing,
sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of
use (“Terms”). For these purposes, Springer Nature considers academic use (by researchers and
students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and
conditions, a relevant site licence or a personal subscription. These Terms will prevail over any
conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription (to
the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of
the Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may
also use these personal data internally within ResearchGate and Springer Nature and as agreed share
it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not otherwise
disclose your personal data outside the ResearchGate or the Springer Nature group of companies
unless we have your permission as detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial
use, it is important to note that Users may not:

1. use such content for the purpose of providing other users with access on a regular or large scale
basis or as a means to circumvent access control;
2. use such content where to do so would be considered a criminal or statutory offence in any
jurisdiction, or gives rise to civil liability, or is otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association
unless explicitly agreed to by Springer Nature in writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a
systematic database of Springer Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a
product or service that creates revenue, royalties, rent or income from our content or its inclusion as
part of a paid for service or for other commercial gain. Springer Nature journal content cannot be
used for inter-library loans and librarians may not upload Springer Nature journal content on a large
scale into their, or any other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not
obligated to publish any information or content on this website and may remove it or features or
functionality at our sole discretion, at any time with or without notice. Springer Nature may revoke
this licence to you at any time and remove access to any copies of the Springer Nature journal content
which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or
guarantees to Users, either express or implied with respect to the Springer nature journal content and
all parties disclaim and waive any implied warranties or warranties imposed by law, including
merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published
by Springer Nature that may be licensed from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a
regular basis or in any other manner not expressly permitted by these Terms, please contact Springer
Nature at

onlineservice@springernature.com