Expert Teachers’Interactive Cognition:
an Analysis of Stimulated Recall Interviews
Abstract
The presented study focuses on the interactive cognition of expert teachers
during their teaching. 16 foreign language teachers’ lessons were videotaped and
the teachers were asked to reveal their interactive cognition through a stimulated
recall interview.The verbal protocols were then analyzed in the light of argumentation
analysis and the claims were subject to content analysis. The results showed
that individual teachers varied greatly as regards their percentages of stimulated
recall as well as other aspects of their interactive cognition, which supports the
prototypical view of teacher expertise.
Keywords: interactive cognition, stimulated recall, pedagogical content knowledge,
foreign language teaching
Introduction
In this paper we focus on the issue of expert teachers’ interactive cognition
while teaching.The data were elicited from 16 expert teachers of foreign languages
(English and German) at Czech lower-secondary schools using stimulated recall
and analyzed in the light of argumentation analysis.Subsequently,content analysis
was conducted, focusing mainly on pedagogical content knowledge.
The analyses were carried out within the research project “Expert teacher: the
nature of expertise and determinants of professional development (in FLT perspective)”,whose
aims include a theoretical and empirical investigation into the nature
of expert foreign language teachers (Píšová et al., in preparation). As regards this
František Tůma, Michaela Píšová,
Petr Najvar, Věra Janíková
Czech Republic
290 František Tůma, Michaela Píšová, Petr Najvar, Věra Janíková
particular study, the method and the results of the pilot study have been briefly
introduced elsewhere (Janíková, Tůma, & Najvar, 2013). In the presented study we
focus on one part of the research, whose aim was to capture some characteristics
of teachers’ interactive cognition during their teaching.
1. Interactive cognition and pedagogical content
knowledge
In the context of this study, we see interactive cognition, in line with Schepens,
Aelterman and van Keer (2007) and others, as a part of teachers’ practical
knowledge. We understand interactive cognition as the teacher’s thoughts while
teaching, which are directly unavailable to researchers, since trying to address
them in-action would result in disturbing of the natural dynamics of the observed
phenomena. Researchers are therefore left to rely on indirect approaches. In
our study we relied on stimulated recall, which we introduce and discuss in the
methodological section.
As regards research into teacher expertise, interactive cognition seems to be
a promising focus. Cognitive research shows that experts, by definition, are able
to think more effectively about problems. Glaser and Chi (1988) define expertise
as the possession of an organised body of conceptual and procedural knowledge
that is readily accessible and can be used with superior metacognitive skills. In
teaching,a highly influential framework for teacher knowledge base was proposed
by Shulman (1986, 1987). Its critical constitutive element, and thus the criterion
defining feature of expertise in teaching,is pedagogical content knowledge (PCK),
i.e. knowledge that guides teachers´ actions and reasoning concerning content to
be taught in highly contextualised classroom settings.
Shulman (1986, 1987) described PCK as including subject matter knowledge,
knowledge of potential student learning difficulties and students’prior knowledge
of specific concepts, as well as the most effective models, analogies, illustrations,
explanations,and considerations to make the content understandable for students.
Later on, Shulman’s framework was elaborated on by numerous researchers, e.g.
Grossman (1990),Magnusson et al.(1999),who attempted to further conceptualise
PCK to identify its components. A significant contribution to theoretical development,
analytical clarification as well as empirical testing of the concept was
made by Ball, Thames and Phelps (2008; content knowledge for teaching), who
focused specifically on teachers of mathematics. For the purposes of this study
PCK represents the core of teachers’ interactive cognition.
291Expert Teachers’Interactive Cognition
2. Methodology
The aim of our study was to analyze the interactive cognition of expert teachers.
Our main research question was: What was the nature of the teachers’ interactive
cognition? Access to interactive cognition can only be gained through language.
In this study,argumentation analysis of stimulated recall and subsequently content
analysis were deployed.
2.1. Argumentation and argumentation analysis
Modern argumentation theory builds on logic and philosophy as well as on
pragmatics, psychology and sociology (Toulmin, 2003). The layout of an argument
has been outlined by Toulmin (2003, pp. 87–134). The two main components of
an argument are claim and data. Whereas the claim (C) is the conclusion of an
argument, the data (D) is a foundation for the claim (Toulmin, 2003, p. 90). The
most typical forms of arguments are “C, because D” or “D, so C” (Toulmin, 2003,
p. 99).
Kopperschmidt (1985, p. 161) defines argumentation analysis as a method that
reconstructs the specific and logical structure of argumentative discourse. This
method has been used in a number of studies,predominantly in the field of science
education (cf., e.g., Kelly, Druker, & Chen, 1998; Kerlin, McDonald, & Kelly, 2010).
For the purposes of this study,argumentation analysis was used for segmenting the
teachers’ utterances into units, some of which were subject to subsequent content
analysis.We introduce the procedure in section 2.4.
2.2. Stimulated recall
In order to access cognitive processes, introspective process tracing methodology
is most frequently utilised based on the tested assumption that a respondent
can verbalise them at some level (Ericsson & Simon, 1999). Out of a number of
methods available, stimulated recall is considered relevant to this study. A stimulated
recall interview requires participants to report what they were thinking and/
or feeling during a past activity.
Without a visual, aural or written artefact reminder, stimulated recall would be
open to charges of non-reliability and non-validity (Ericsson & Simon, 1999). In
response,Henderson and Tallman (2006) proposed six issues that need consideration
in order to strengthen reliability and validity: (1) timing of the stimulated
recall interview,(2) capturing the data,(3) conduct of the interview,(4) interviewer
training, (5) interviewee and interviewer fatigue, and (6) coding and categorizing
the data. In our research we paid special attention to these issues in order to
292 František Tůma, Michaela Píšová, Petr Najvar, Věra Janíková
maximise the reliability and validity of our data. For details cf., Píšová et al. (in
preparation).
2.3. Data collection
Lessons of 16 expert teachers1
of foreign languages (English and German) at
Czech lower-secondary schools were video recorded.Immediately after each lesson,
the teacher was asked to comment on a 20-minute part of the video. The teachers
were asked to stop the video every time they recalled what they had been thinking
about. The instructions by the researchers included an explicit statement that the
purpose of the session was to recall what the teacher had been thinking about at the
moment depicted in the video (interactive cognition),not at the moment of seeing
the video (reflection).The researcher only listened and could ask questions in order
to guide the teacher towards stimulated recalling (e.g.,“Is this what you are thinking
about now or what you were thinking at that moment during your teaching?”).
The session lasted approximately 45 minutes and its audio recording was subsequently
transcribed.The transcription included contextualization cues (Gumperz,
1992) captured in the recording,mainly paralinguistic features (e.g.,laughter,pause,
hesitation). The average length of a transcript was 2904 words. The transcripts of
each session were subject to subsequent analysis.
2.4. Data analysis
The analysis was guided by the main research question: What was the nature
of the teachers’ interactive cognition?, which was broken down into the following
subquestions:What was the proportion of claims related to thinking while teaching?,
What was the proportion of justifications of these claims?,What was the proportion
of recalled claims related to thinking about teaching?,and In what categories (PCK
components) were the claims related to thinking in teaching distributed?
The transcripts of the stimulated recall interviews were analyzed with regard to
each teacher’s way of stimulated recall (e.g., most teachers used the past tense for
recalled claims, whereas some used the present tense). The identification of claims
recalled by each teacher also included (1) the character of the utterance (a typical
claim was relatively short and related to one action in the video recording), (2)
contextualization cues (Gumperz, 1992) and (3) a complex understanding of the
1
These 16 expert teachers constituted the sample for the second phase of the research
project within which stimulated recall was carried out. The sampling at the initial stage was
purposive and out of the 30 teachers,16 teachers were selected for the second phase on the basis
of their classroom performance data. Cf., Píšová et al. (in preparation) for details.
293Expert Teachers’Interactive Cognition
teachers’ utterances.A similar way of identifying argument structure was used by
other researchers (e.g., Kelly et al., 1998).
The analysis comprised two steps.First,the teachers’discourse was segmented in
the light of the argumentation analysis.Second,the claims were subject to quantitative
content analysis.
During the segmentation, we first identified the core of a recalled utterance, i.e.,
the recalled claim which was related to the teacher’s interactive cognition. Second,
we identified the supporting data, which we marked as justification of the claim.
Next, the utterances related to the teachers’ reflection were coded accordingly.
Other utterances were excluded from the analysis.
The recalled claims, i.e., utterances related to the teachers’interactive cognition,
were further investigated by means of content analysis.Each claim was,in line with
a synthesis of the conceptualizations and models of PCK (cf., section 1), coded
into one of the following core categories: processes,content and aims.Within each
of these categories a distinction was made between claims oriented to learners
or subject matter as suggested by Janík et al. (2009), Píšová et al. (2011). Having
piloted the procedure (Janíková et al.,2013),we realized that although the majority
of claims fitted the category system for PCK, there was a group of claims which
were related to the teachers’ emotions during their teaching. The emergence of
emotions in cognitively oriented research in a way supports the validity of the
analysis: it is generally acknowledged that cognition is shaped by emotions.Moreover,
teaching has been recognized as an inherently emotional enterprise both in
the classroom and outside (for reviews, cf., Hargreaves, 1998; Sutton & Wheatley,
2003). Therefore, we added another dimension to our category system. The final
version of the category system accompanied by examples2
can be found in Table
1. The italics in the examples indicate the recalled claim, whereas the rest of each
utterance is the justification of the claim.
Table 1. Category system for content analysis of recalled claims
Main category
Orienta-
tion
Example
Cognitive
dimension
PCK – processes (the how
of teaching – strategies,
techniques, methods, tasks,
etc.)
Pupils “Here I intentionally chose Jana, because
she is the weakest … in order to check
whether she knew this.”
Subject
matter
“Here I realized that it would be beneficial
to go back to the task, because there was a
lot of new vocabulary.”
2
The examples were translated from Czech.
294 František Tůma, Michaela Píšová, Petr Najvar, Věra Janíková
Main category
Orienta-
tion
Example
Cognitive
dimension
PCK – content (the what
of teaching)
Pupils “Here I used the word ‘blush’ and I thought
that the majority of them would not know
the word, so I put the word to a sentence,
so that those who are ready to remember
the word could link it with the meaning.”
Subject
matter
“Here I must admit that the word ‘trolley’
slipped my memory, but usually before
I get to the board I recall the word, or I
have a dictionary at hand.”
PCK – aims (the why of
teaching)
Pupils “Here I intentionally said ‘morning’ in
Czech, so that they [the learners] could
compare the meaning in Czech and
English.”
Subject
matter
“Here I thought that it was more important
for the activity in that moment that
they discuss it with each other [no matter
in what language – Czech or English] and
that it will be sufficient if they formulate
their conclusions in English.”
Affective
dimension
Expressing emotions or
mood
- [after not being able to start a video on an
interactive board] ”I was angry, because
the kids were waiting and nothing was
going on.”
4. Results
Having piloted and adjusted the category system (for details cf., Janíková et al.,
2013) we reached a relatively satisfactory direct agreement among three researchers
(76% on identifying claims, 82% on identifying justification, 93% on identifying
reflection and 60% on the content analysis).Evidently,some aspects were relatively
straightforward to code,whereas some utterances required discussion in the team.
Therefore, each of the transcripts was coded by one researcher and the areas of
difficulty, especially coding according to the category system, were subsequently
discussed by the group of three coders. Furthermore, to assure consistency in
coding, each transcript was checked by another researcher who had not coded it.
From our theoretical positions and the description of the category system for
content analysis (Table 1) it follows that some categories were complementary to
others (stimulated recall – reflection, orientation toward learners – orientation
295Expert Teachers’Interactive Cognition
to subject matter), whereas some categories (affect, justification of claims) were
autonomous. In order to answer the research questions stated above, the following
variables were defined.All of the variables were computed separately for each
teacher.
The percentage of recalled claims was defined as the ratio of the number of
recalled claims to the sum of numbers of recalled claims and utterances coded as
reflection.
The percentage of reflection was defined as the opposite of the percentage of
recalled claims – it is the proportion of the number of utterances marked as reflection
to the sum of recalled claims and utterances marked as reflection.
The percentage of justification of claims was defined as the ratio of the number
of justifications to the total number of recalled claims.
The percentage of affect was defined as the proportion of claims classified as
affective to the total number of recalled claims.
The orientation toward learners (or subject matter) was defined as the proportion
of the sum of the recalled claims classified as oriented to learners (or subject
matter) to the total number of recalled claims. Like the other variables, these two
variables were expressed in percentage.
The analysis revealed striking differences between individual teachers as regards
the numbers of recalled claims and utterances classified as reflection. As regards
the distribution of recalled claims, out of the total of 247 recalled claims, the vast
majority were classified as PCK – processes (184 claims), relatively fewer were
classified as PCK – aims (11 claims) and as PCK – content (11 claims). 41 claims
were classified as related to affect. The results for individual teachers are shown
in Table 2.
Table 2. Results for individual teachers
Teacher
The sum
of recalled
claims and
utterances
marked as
reflection
Percentage
of
recalled
claims
[%]
Percentage
of
reflection
[%]
Orientation
to
learners
[%]
Orientation
to
subject
matter
[%]
Percentage
of
affect
[%]
Percentage
of
justifica-
tion
[%]
T01 46 22 78 80 20 0 60
T02 22 86 14 74 11 16 74
T03 16 81 19 31 62 8 92
T04 33 79 21 54 19 27 62
T05 51 25 75 54 46 0 77
296 František Tůma, Michaela Píšová, Petr Najvar, Věra Janíková
Teacher
The sum
of recalled
claims and
utterances
marked as
reflection
Percentage
of
recalled
claims
[%]
Percentage
of
reflection
[%]
Orientation
to
learners
[%]
Orientation
to
subject
matter
[%]
Percentage
of
affect
[%]
Percentage
of
justifica-
tion
[%]
T06 35 94 6 45 36 18 67
T07 48 8 92 25 50 25 75
T08 42 69 31 48 17 34 97
T09 33 55 45 78 11 11 89
T10 19 74 26 86 7 7 50
T11 22 82 18 61 11 28 50
T12 14 36 64 60 20 20 80
T13 35 31 69 64 27 9 27
T14 27 59 41 100 0 0 75
T15 28 32 68 67 11 22 44
T16 23 39 61 67 22 11 33
5. Discussion on results and conclusions
To address the main research question – What was the nature of the teachers’
interactive cognition?- we can conclude that the teachers overall recalled mainly
their cognition related to PCK – processes (184 out of 247 claims),which might be
interpreted as a natural consequence of the elicitation.The teachers saw themselves
teaching, and with regard to the nature of the task of recalling what they had
been thinking about, the considerations behind their actions, teaching strategies,
techniques and methods seemed to be the most accessible aspects of their performances.
It may, however, document a more general feature of teachers’ thinking,
specifically the fact that teachers tend to focus more on the procedural aspects of
their work as noted by Janík, Knecht, Najvar et al. (2010), or Janík, Janko, Knecht
et al. (2010) in their research into curricular reform implementation in the Czech
Republic.
However, the extent to which individual teachers revealed their interactive cognition
varied greatly.As the results showed, there were teachers whose percentage
of recalled claims was very high (e.g., T06 with 86%) and there were teachers
who were, despite the researchers’ prompts and questions, rather reflective (e.g.
T07 whose percentage of reflection was 92). The results for the latter group, i.e.,
297Expert Teachers’Interactive Cognition
for the teachers whose utterances were mainly reflective, should be interpreted
with caution, since the orientation to pupils or the subject matter is computed on
the basis of their recalled claims. The data also revealed that the teachers differed
from each other in the light of the other variables. It can be concluded that the
group of 16 teachers did not seem to share a common feature in the light of the
results. Instead, there seem to be certain areas of expertise which are developed to
different extent in each individual or are displayed in response to the particular
pedagogical situation. This seems to be compatible with the prototype view of
expertise (Sternberg & Horvath, 1995).
It should be pointed out that the fact that some teachers did not reveal much
of their interactive cognition through stimulated recall, and instead were rather
reflective, does not mean that they were better or worse teachers. It follows from
our previously collected data that all 16 teachers displayed some characteristics of
expertise. One interpretation of the high percentage of reflection in some teachers
(especially T07, T01 and T05) can be that they found it more difficult than
others to verbalize their cognitive processes.Another interpretation can be that in
the expert teachers such cognitive processes were routinized and automatized to
such an extent that they were unable to verbalize them. This is in accord with the
findings of cognitive psychology research on memory traces, more specifically on
the distinction between memory traces in novices and experts (e.g., Ericsson &
Simon, 1999).
The data were segmented by means of argumentation analysis and subsequently
analyzed in the light of a category system based on PCK.The argumentation analysis
seemed suitable for this type of data and allowed for quantifying the recalled
claims and their justifications.
The presented study has its limitations. The way the data was elicited and analyzed
might have limited the findings. Contrary to our initial aim to concentrate
exclusively on interactive cognition,reflection-on-action occurred in many verbal
protocols. Also, the generalizability of the findings is limited. The data obtained
in the stimulated recall only relate to the lesson that the teachers had taught. The
content of the lesson as well as the specific group of learners might have influenced
the flow of the lesson, and thus the teacher’s interactive cognition. Relatedly, it
should be pointed out that our results are domain specific (teaching English or
German) and the research was conducted in the cultural context of the Czech
Republic.
The results presented here comprise one stage of a research project which was
conducted as a multiple case study. For individual cases (teachers), multiple data
sources were used, and a combination of data analysis techniques was applied.
298 František Tůma, Michaela Píšová, Petr Najvar, Věra Janíková
Apart from the analysis presented here, cluster analysis was conducted and the
transcripts were analyzed inductively in order to reveal other aspects of the stimulated
recall (Píšová et al., in preparation).
Acknowledgement: This study was supported by grants GA ČR P407/11/0234 “Expert
Teacher: the nature of expertise and determinants of professional development (in FLT
perspective)” and CZ.1.07/2.3.00/30.0009 “Employment of newly graduated doctors of
science for scientific excellence“.
References
Ball,D.,Thames,M.H.,& Phelps,G. (2008).Content Knowledge for Teaching:What
Makes It Special? Journal of Teacher Education, 59(5), 389–407.
Ericsson, K.A., & Simon, H.A. (1999). Protocol analysis: Verbal reports as data (3rd
ed.). Cambridge: MIT Press.
Glaser, R., & Chi, M. (1988). Overview. In M. Chi, R. Glaser, & M. Farr (Eds.), The
nature of expertise (pp. xv–xxvii). Hillsdale: Erlbaum.
Grossman, P.L. (1990). The making of a teacher. Teacher knowledge and teacher
education. New York: Teachers College Press.
Gumperz, J.J. (1992). Contextualization and understanding. In A. Duranti &
C. Goodwin (Eds.),Rethinking context. Language as an interactive phenomenon.
(pp. 229–252). Cambridge: Cambridge University Press.
Hargreaves, A. (1998). The emotional practice of teaching. Teaching and Teacher
Education, 14(8), 835–854.
Henderson, L., & Tallman, J. (2006). Stimulated recall and mental models: Tools for
teaching and learning computer information literacy.Lanham: The Scarecrow Press.
Janík, T., Maňák, J., & Knecht, P. (2009). Cíle a obsahy školního vzdělávání a
metodologie jejich utváření. Brno: Paido.
Janík, T., Najvar, P., Slavík, J., & Trna, J. (2009). On the dynamic nature of physics
teachers pedagogical content knowledge. Orbis scholae, 3(2), 47–62.
Janík,T.,Knecht,P.,Najvar,P.,Pavlas,T.,Slavík,J.,& Solnička D. (2010).Kurikulární
reforma na gymnáziích v rozhovorech s koordinátory pilotních a partnerských škol.
Praha: Výzkumný ústav pedagogický v Praze.
Janík,T.,Janko,T.,Knecht,P.,Kubiatko,M.,Najvar,P.,Pavlas,T.,Slavík,J.,Solnička,
D.,& Vlčková,K. (2010).Kurikulární reforma na gymnáziích: výsledky dotazníkového
šetření.. Praha: Výzkumný ústav pedagogický v Praze.
299Expert Teachers’Interactive Cognition
Janíková,V., Tůma, F., & Najvar, P. (2013). Fremdsprachenlehrkräfte als Experten:
Teilergebnisse einer introspektiven Studie. In V. Janíková & R. Seebauer (Eds.),
Education and Languages in Europe. Bildung und Sprachen in Europa (pp.
121–130). Wien/Berlin: LIT Verlag.
Kelly, G.J., Druker, S., & Chen, C. (1998). Students’ reasoning about electricity:
combining performance assessments with argumentation analysis.International
Journal of Science Education, 20(7), 849–871.
Kerlin, S.C., McDonald, S.P., & Kelly, G.J. (2010). Complexity of Secondary Scientific
Data Sources and Students’Argumentative Discourse.International Journal
of Science Education, 32(9), 1207–1225.
Kopperschmidt, J. (1985). An analysis of argumentation. In T.A. van Dijk (Ed.),
Handbook of discourse analysis. Volume 2. (pp. 159–168). London: Academic
Press.
Magnusson, S., Krajcik, J., & Borko, H. (1999). Nature, sources and development
of pedagogical content knowledge for science teaching. In J. Gess-Newsome &
N.G. Lederman (Eds.), Examining pedagogical content knowledge: The construct
and its implications for science education (pp. 95–132). Dordrecht: Kluwer
Academic.
Píšová, M., Kostková, K., & Janík, T. (Eds.) (2011). Kurikulární reforma na gymnáziích:
případové studie tvorby kurikula. Praha: Výzkumný ústav pedagogický.
Píšová, M. et al. (in preparation). Učitel expert. Výzkumná zpráva. [working title].
Brno: Masarykova univerzita.
Shulman, L.S. (1986). Paradigms and research programs in the study of teaching.
A contemporary perspective. In M.C. Wittrock (Ed.), Handbook of research on
teaching (pp. 3–36). New York: MacMillan.
Shulman, L.S. (1987). Knowledge and teaching. Foundations of the New Reform.
Harvard Educational Review, 57(1), 1–22.
Schepens,A.,Aelterman,A., & van Keer, H. (2007). Studying learning processes of
student teachers with stimulated recall interviews through changes in interactive
cognitions. Teaching and Teacher Education, 23(4), 457–472.
Sternberg, R.J., & Horvath, J.A. (1995).A prototype view of expert teaching. Educational
Researcher, 24(6), 9–17.
Sutton,R.E.,& Wheatley,K.F. (2003).Teachers’emotions and teaching: a review of
the literature and directions for future research. Educational Psychology Review,
15(4), 327–358.
Toulmin, S.E. (2003). The uses of argument. Cambridge: Cambridge University
Press.