CC BY-NC-ND
0
Psychology. Journal of the Higher School of Economics. 2024. Vol. 21. N 2. P. 408-422. Психология. Журнал Высшей школы экономики. 2024. Т. 21. № 2. С. 408-422. DOI: 10.17323/1813-8918-2024-2-408-422
INSIGHT PROBLEM OR INSIGHTFUL SOLUTION: IS IT STILL WORTH USING INSIGHT PROBLEMS TO INVESTIGATE INSIGHT?
A.D. SAVINOVAa, N.YU. LAZAREVAa, A.V. CHISTOPOLSKAYAa, I.N. MAKAROVa, S.YU. KOROVKINa, I.YU. VLADIMIROVa
aP.G. Demidov Yaroslavl State University, 14 Sovetskaya Str, Yaroslavl, 150000, Russian Federation
Инсайтная задача или инсайтное решение: стоит ли использовать инсайтные задачи для изучения инсайта?
А.Д. Савинова", Н.Ю. Лазарева", А.В. Чистопольская", И.Н. Макаров", С.Ю. Коровкин", И.Ю. Владимиров"
'Ярославский государственный университет им. П.Г. Демидова, 150000, Россия, Ярославль, ул. Советская, д. 14
Abstract
Traditional insight studies assumed that there is a special class of problems called "insight problems" whose solution would cause feelings of insight. However, it has been previous shown that insight problems can be solved in both insightful (with Aha! experience) and step-by-step (without Aha! experience) ways, and exactly the same is observed for analytical problems. The present work addresses the question whether it is sufficient to use the problem types to detect insightful feelings or whether it is necessary to use the solution types. For this purpose we collected the dataset of previously published open data which used Danek and Wiley's questionnaire
Резюме
Традиционные исследования инсайта предполагают, что существует особый класс так называемых инсайтных задач, чье решение приводит к переживанию инсайта. Однако было показано, что инсайтные задачи могут быть решены как инсайтным (с ага-переживанием), так и пошаговым путем (без ага-пережива-ния), и точно такая же ситуация наблюдается для решения аналитических задач. Данная работа посвящена вопросу, действительно ли достаточно использовать типы задач для определения инсайтных переживаний или же необходимо использовать тип решения. Для этой цели мы собрали базу ранее опубликованных открытых данных, использующих опросник А. Данек и Дж. Вайли с различными
The research was supported by Russian Science Foundation, grant 22-18-00358, https://rscf.ru/en/project/22-18-00358/
Исследование поддержано Российским научным фондом, проект № 22-18-00358, https://rscf.ru/project/22-18-00358/
for subjective ratings of insight phenomenology. Joint analysis showed that using solution types provided greater benefits for insight investigation than using problem types: insightful solutions were more pleasant, sudden, and relieving than step-by-step solutions, but insight and analytical problems differed from each other on only suddenness scale. We concluded that relying only on the problem type is not the best strategy for investigation of insight, because insight and analytical problems are quite similar in terms of the Aha! experience. The use of solution types is much more fruitful and distinguishes various situations in the process of problem solving. The obtain results were inter-pretated in the context of relationships between the Aha! experience and representational change.
Keywords: insight, Aha! experience, phenomenology, insightful solution, step-by-step solution, subjective rating, joint analysis
Anna D. Savinova — Research Fellow, Laboratory for Cognitive Research, De-midov Yaroslavl State University, PhD in Psychology.
Research Area: experimental psychology, cognitive science, problem solving, thinking, insight, working memory, Aha! experience.
E-mail: anuta1334@yandex.ru
Natalia Yu. Lazareva — Junior Research Fellow, Laboratory for Cognitive Research, Demidov Yaroslavl State University, PhD in Psychology. Research Area: experimental psychology, cognitive science, problem solving, thinking, insight, working memory, Aha! experience.
E-mail: lazareva natasha93@mail.ru
Alexandra V. Chistopolskaya — Associate Professor, Department of General Psychology, Faculty of Psychology,
субъективными шкалами для оценки феноменологии инсайта. Совместный анализ показал, что использование типа решения дает больше преимуществ для изучения инсайта, чем использование типов задач: инсайтные решения были более приятными, внезапными и облегчающими, чем пошаговые решения, при этом инсайтные и аналитические задачи отличались друг от друга только по шкале внезапности. Мы пришли к выводу, что полагаться исключительно на тип задач - не лучшая стратегия для изучения инсайта, поскольку инсайт-ные и аналитические задачи достаточно похожи в терминах ага-переживания. Использование типов решения гораздо более плодотворно и позволяет различать разнообразные ситуации в процессе решения задач. Полученные результаты были проинтерпретированы в контексте взаимоотношений между ага-переживанием и изменением репрезентации.
Ключевые слова: инсайт, ага-переживание, феноменология, инсайтное решение, пошаговое решение, субъективные шкалы, совместный анализ.
Савинова Анна Джумберовна — научный сотрудник, лаборатория когнитивных исследований, Ярославский государственный университет им. П. Г. Демидова, кандидат психологических наук.
Сфера научных интересов: экспериментальная психология, когнитивная наука, решение задач, мышление, инсайт, рабочая память, ага-переживание.
Контакты: anuta1334@yandex.ru
Лазарева Наталья Юрьевна — младший научный сотрудник, лаборатория когнитивных исследований, Ярославский государственный университет им. П. Г. Демидова, кандидат психологических наук. Сфера научных интересов: экспериментальная психология, когнитивная наука, решение задач, мышление, инсайт, рабочая память, ага-переживание.
Контакты: lazareva_natasha93@mail.ru
Чистопольская Александра Валерьевна —
доцент, кафедра общей психологии, факультет психологии, Ярославский государственный
Demidov Yaroslavl State University, PhD in Psychology.
Research Area: cognitive psychology, insight problem solving, insight, embodiment cognition, psychology of art. E-mail: chistosasha@mail.ru
Igor N. Makarov — Engineer, Laboratory for Cognitive Research, Demidov Yaroslavl State University. Research Area: experimental psychology, cognitive science, problem solving, thinking, insight. E-mail: reoge@mail.ru
Sergei Yu. Korovkin — Head of Laboratory, Professor, Laboratory for Cognitive Research; Department of General Psychology, Faculty of Psychology, Demidov Yaroslavl State University, DSc in Psychology.
Research Area: experimental psychology, cognitive science, problem solving, thinking, insight, working memory, Aha! experience.
E-mail: korovkin_su@list.ru
Ilia Yu. Vladimirov — Associate Professor, Department of General Psychology, Faculty of Psychology, Demidov Yaroslavl State University, PhD in Psychology.
Research Area: experimental psychology, cognitive science, problem solving, thinking, insight.
E-mail: kein17@mail.ru
университет им. П. Г. Демидова, кандидат психологических наук.
Сфера научных интересов: когнитивная психология, решение инсайтных задач, инсайт, воплощенное познание, психология искусства. Контакты: chistosasha@mail.ru
Макаров Игорь Николаевич — инженер, лаборатория когнитивных исследований, Ярославский государственный университет им. П.Г. Демидова.
Сфера научных интересов: экспериментальная психология, когнитивная наука, решение задач, мышление, инсайт. Контакты: reoge@mail.ru
Коровкин Сергей Юрьевич — руководитель, профессор, лаборатория когнитивных исследований, кафедра общей психологии, факультет психологии, Ярославский государственный университет им. П. Г. Демидова, доктор психологических наук.
Сфера научных интересов: экспериментальная психология, когнитивная наука, решение задач, мышление, инсайт, рабочая память, ага-переживание.
Контакты: korovkin_su@list.ru
Владимиров Илья Юрьевич — доцент, кафедра общей психологии, факультет психологии, Ярославский государственный университет им. П. Г. Демидова, кандидат психологических наук.
Сфера научных интересов: экспериментальная психология, когнитивная наука, решение задач, мышление, инсайт. Контакты: kein17@mail.ru
For many years, one of the central issues of the psychology of thinking has been the search for specific characteristics of insight. The studies of insight are usually associated with the use of a special problem type — insight problems. According to the Representational Change Theory (RCT), the main difference between insight and analytical problems is that the latter do not require a representational change, which occurs when the initial problem representation is incorrect and does not allow the answer to be found (Ohlsson, 1992). The result of a successful representational change can be an Aha! experience that is described as a pleasant feeling of ",seeing' the complete solution in the mind's eye" (Ibid., p. 5). Thus, it is assumed that the Aha! experience can only occur in insight problems. Meanwhile, it has
become evident that insight problems can sometimes be solved in a step-by-step way, i.e., without the Aha! experience (Weisberg, 2015; Danek et al., 2016; Webb et al., 2016, 2018), and analytical problems can be solved with the Aha! experience (Webb et al., 2018; Lazareva & Vladimirov, 2019).
However, several questions arise at this point. Firstly, suppose that the Aha! experience is strictly associated with and occurs only after a successful representational change. Is it possible to say that a representational change can only occur in insight problems? It is much more logical to assume that — due to the specific participants' knowledge or due to preliminary experimental influences — "analytical problem" may involve a representational change, i.e., it becomes insightful and is solved with the Aha! experience. In this case, it makes no sense to talk about insight problems, rather than about insightful solutions. Another implication is that insight studies should not be limited to insight problems, they can include a wide variety of experimental material containing a representational change (jokes, modern arts, recognition of noisy or double images, etc.).
This is not a new question, but many authors still use insight problems to study insight (Chuderski, Jastrz^bski, & Kucwaj, 2021; Korovkin et al., 2021; Salmon Mordekovich & Leikin, 2022), because this problem type has a wide theoretical background and explanation why pure insight problems can cause the Aha! experience (e.g., Ohlsson, 1992, 2011; Weisberg, 1995, 2015). From this point of view, it is not clear why analytical problems can be solved by insightful strategy, because they do not have all the necessary features. This adherence does not allow us to dot the i's on the issue of using problem types instead of switching to solution types.
Secondly, suppose that the Aha! experience is not strictly associated with a representational change, and then even deeper methodological questions arise: What is more important for investigation of insight — the presence of a representational change or the presence of the Aha experience? Can a solution be considered as insightful if it does not involve the Aha! experience? What causes the Aha! experience? We will not devote this paper to attempts to answer these questions, but they are actively discussed and studied in modern works (e.g., Ammalainen & Moroshkina, 2021; Becker et al., 2020; Becker et al., 2021; Danek et al., 2020).
Even if we assume that the Aha! experience is strictly associated with a representational change, we are faced with a contradiction. On the one hand, there is RCT, which posits that problem types cause certain metacognitive feelings (the Aha! experiences) due to the presence or absence of a representational change in the solution (Ohlsson, 1992, 2011). On the other hand, there is some empirical evidence showing that the problem type does not always uniquely predict the presence of the Aha! experience in a solution (Danek et al., 2016; Webb et al., 2018; Lazareva & Vladimirov, 2019). In this regard — with a help of joint analysis — we want to check how often insight and analytical problems are accompanied by the Aha! experience. We also want to know whether it makes sense to completely abandon the dichotomy of problem types in favor of using solution types: insightful solutions for cases with the Aha! experience and step-by-step solutions for cases without the Aha! experience.
Thus, two contradictory hypotheses can be constructed:
1. According to RCT, insight and analytical problems generally differ from each other. Insight problems score higher on the most subjective scales of the Aha! experience compared to analytical problems.
2. According to previous empirical data, problem types do not differ from each other, but there are differences between insightful and step-by-step solutions. Insightful solutions are evaluated higher on the most subjective scales of the Aha! experience than step-by-step solutions.
Method
To test our ideas, we decided to conduct joint analysis of previous studies, because it provides two significant advantages: a large sample size and — inasmuch as used studies were conducted in different countries — control of such factors as the influence of an experimenter, translation of a questionnaire, and cross-cultural differences. Post-experimental self-reports are commonly used to measure the Aha! experience, and there are different variants of them (Novick & Sherman, 2003; Jung-Beeman et al., 2004; Bowden et al., 2005; Wong, 2009). Currently, the questionnaire on the Aha! phenomenology (Danek et al., 2014; Danek & Wiley, 2017) seems to have received the greatest prevalence and can be considered as the most developed and tested on various problem types, because a substantial pool of experimental data has been accumulated. According to Danek and Wiley's (2017), the Aha! experience is understood as a multi-dimensional construct that consists different aspects: pleasure, surprise, suddenness, relief, certainty and drive. The dataset was collected from studies where this questionnaire was used to evaluate the solvers' subjective experiences.
Selection of Studies
Using Google Scholar we analyzed published studies that cited Danek and Wiley (2017). At the time of dataset creation (January 2022), 114 papers were found. We selected papers with the data publicly available and full, not aggregated data. We used only empirical papers and papers without priming or hints in experimental design. Only control groups were included in the joint analysis. Thus, data were obtained from 9 studies, including 9,274 rows of data or 990 unique participants. It is worth noting that the available data were very heterogeneous, so they required some changes or exclusions for the unity of the final dataset. Clarifications of excluded data can be found in OSF at https://osf.io/j6s38/. The full list of studies is presented in Table 1.
Characteristics of the Dataset
A number of characteristics in the collected dataset were identified and each data row was evaluated for compliance with them. The following characteristics were used:
Table 1
List of the Studies Included
Reference Sample Problem types Available scales Excluded data
Bilalic et al., 2021 74 participants (59 females), aged 1762 years (M = 24, SD = 6.1) Classical insight problems Pleasure1, surprise2, sudden- ness3, certainty4 Participants with no experience in playing chess were taken. Besides, we used successful solutions, because participants evaluated subjective ratings only after successful solution attempts
Chuderski, Jastrzçbski, & Kucwaj, 2021 200 participants (139 females), aged 18-38 years (M = 23.5, SD = 4.3) Classical insight problems Pleasure, suddenness, relief5, certainty
Chuderski Jastrzçbski, Kroczek et al., 2021 100 participants (73 females), aged 1832 years (M = 22.6, SD = 2.94) Analytical problems, classical insight problems, Raven matrices (as analytical problems) Pleasure, suddenness, certainty
Danek & Wiley, 2017 70 participants (48 females), age range is not specified (M = 19.6, SD = 2.8) Magic tricks Pleasure, surprise, suddenness, relief, certainty, drive
Danek & Wiley, 2020 127 participants (56 females), aged 1742 years (M = 19.04, SD = 2.1) Magic tricks Pleasure, surprise, suddenness, relief, certainty, drive6 We used only the Experiment 2, because in the Experiment 1 only two subjective parameters were measured: overall aha! experience and certainty of the solution
1 At the moment of solution, my feelings were... (unpleasant — pleasant).
2 The moment of solution was... (not surprising — surprising).
3 This solution came to me... (in steps — all at once).
4 How certain are you that your solution is correct: (uncertain — certain).
5 At the moment of solution, I felt... (tense — relieved).
6 I am looking forward to the next problem... (no — yes).
Table 1 (end)
Drqzyk et al., 2020 279 participants (173 females), aged 17-51 years (M = 21.75, SD = 4.42) Analytical problems, classical insight problems Pleasure, suddenness, relief, certainty We used only the control group of Experiment 2, because it contained scales of Danek & Wiley's questionnaire, and the control group did not include additional factor as the ego-depletion
Korovkin et al., 2021 Experiment 1: 40 people (31 females), aged 12-72 years (M = 34.83, SD = 18.43) Experiment 2: 40 people (28 females), aged 14-63 years (M = 30.05, SD = 14.26) Classical insight problems Pleasure, surprise, suddenness, relief, certainty, drive
Spiridonov et al., 2021 40 people (35 females), age range is not specified (M = 19.1, SD = 1.4) CRA Pleasure, surprise, suddenness, relief, certainty, drive We used only the data of Experiment 2
Vladimirov et al., 2021 20 people (17 females), aged 18-21 (M = 20.5, SD = 1.6) Analytical problems, classical insight problems Pleasure, surprise, suddenness, relief, certainty, drive
• Experiment: it indicates authors of the original article of the obtained data.
• Problem type: analytical problems, compound remote associate problems (CRA), classical insight problems, magic tricks, Raven matrices.
• Solution type. Some authors assume that problems can be solved by the insightful way or a step-by-step way. The insightful solution is a solution with the representational change, which is accompanied by specific feelings and metacogni-tive experiences (an Aha! experience, frustration from an impasse, etc.). The step-by-step, or analytical, solution is a solution without the representational change and special feelings, but it is performed through successive steps. In some cases, authors indicate the solution type based on the answer given by the participant. For example, in Korovkin and colleagues' experiment (2021) participants who gave a star-shaped answer were assigned to the insightful solution group, and any other answers were assigned to the step-by-step group. However, the most of works are based on an additional scale assessment — the Aha! experience scale (Danek &
Wiley, 2017; Danek et al., 2020), strategy scale (Chuderski, Jastrz^bski, Kroczek et al., 2021), enlightenment scale (Vladimirov et al., 2021) or insightfulness scale (Spiridonov et al., 2021). In our opinion, all of these scales are similar and are designed to find out how a participant has solved the problem. We used all of this information to determine the solution type, insightful or step-by-step.
• Subjective rating scales: pleasure, surprise, suddenness, relief, certainty and drive (Danek & Wiley, 2017).
• Solved. We indicated how the problem was solved. There are two options: "yes" — successfully solved by the participant, or "no " — solved by the experimenter who gave the correct answer to the participant.
These characteristics are described in more detail at https://osf.io/j6s38/.
A Unified System of Subjective Ratings
The studies differed in the types of scale used for subjective ratings. Some of the studies used a scale from 0 to 100 points (for example, Danek & Wiley, 2017; Danek et al., 2020); others used a scale from 1 to 19 (for example, Dr^zyk et al., 2020) or a scale from 1 to 4 points (Vladimirov et al., 2021). However, this variation was unsuitable for statistical analyses, and all points were transformed to z-scores taking into account the characteristic of "Experiment".
Design & Statistical Analysis
We wanted to investigate how insight and analytical problems, as well as insightful and step-by-step solutions, are different in terms of subjective feelings of the Aha! experience. Different sets of dependent and independent variables were used. The dependent variables were the subjective ratings of the questionnaire (pleasure, surprise, suddenness, relief, certainty, drive). The independent variables were the problem type and the solution type. The data that support the findings of this study are available in OSF at https://osf.io/j6s38/. The data were also derived from the public domain.
Results
Comparison of insight and analytical problems
This section presents results regarding the question whether insight and analytical problems generally differ from each other, i.e., the impact of the problem type on a subjective solution experience. Insight problems (only classical insight problems), analytical problems and Raven matrices (analytical problems and Raven matrices were combined into one group which was labeled as "analytical") were compared. Only successfully solved problems were used (see Figure 1) (N = 2751).
We used Linear Mixed Models with the method of Satterthwaite test. Subjective scales were used as dependent variables, the problem type was used as a fixed effect variable, and the characteristic "Experiment" as a random effect grouping factor. All
Figure 1
Comparison of Insight and Analytical Problems
• Analytical problem A Insight problem
* \ J i I * I i
i T
T-1-1-1-1-r
Pleasure Surprise Suddenness Relief Certainty Drive
Note. Vertical bars denote 95% confidence interval (CI). Significant differences are marked in black, insignificant in grey.
analyses were computed using JASP, v. 0.18.3.0. Insight problems are evaluated as more sudden than analytical problems, t(1, 7.96) = -9.75, p < .001, VS-MPR = 3011.3 (P = -0.255, SE = 0.026). However, we did not find any significant differences in remaining scales.
Comparison of Insightful and Step-by-Step Solutions
The comparison of insightful and step-by-step solutions was carried out for the data containing the labels "insight" and "step-by-step" in the "Solution type" column. Only successfully solved problems were used (see Figure 2) (N = 4,717). We used Linear Mixed Models with the method of Satterthwaite test. Subjective scales were used as dependent variables, the solution type was used as a fixed effect variable, and the characteristic "Experiment" as a random effect grouping factor.
The insightful solution has a higher point on the pleasure scale than the step-by-step solution, t(1, 8.97) = 2.28, p = .048, VS-MPR = 2.5 (P = 0.206, SE = 0.090). The insightful solution is also evaluated higher on suddenness, t(1, 8.67) = 3.22, p = .011, VS-MPR = 7.4 (P = 0.242, SE = 0.075), and relief, t(1, 6.88) = 2.37, p = .050, VS-MPR = 2.4 (P = 0.191, SE = 0.081).
Relationship between the Problem Type and the Solution Type
A Chi-square test of independence was performed to examine the relation between the problem type (insight and analytical) and the solution type (insightful
A.M. CaeuHoea u dp. MHcaumHan 3adaua urn uHcaummepeweHue
417
Figure 2
Comparison of Insight and Step-by-Step Solutions
0.4-
0.2-
a>
i_
o o u>
N 00"
-0.2-
Note. Vertical bars denote 95% confidence interval (CI). Significant differences are marked in black, insignificant in grey.
and step-by-step). Only successfully solved problems were used. The relation between these variables was significant, x2(1, 2664) = 702.2, p < .001. Insight problems were more likely than analytical problems to be labeled as insightful solutions (see Table 2). However, there are many cases when insight problems were solved step-by-step (29.3%), and analytical problems were solved by insight (19.6%).
Discussion
In this work, we wanted to find whether it is enough to use problem types to detect insightful feelings or it is necessary to use solution types. To test this idea, we collected a dataset of previous studies and conducted a joint analysis. The dataset included measurements of the Aha! experience in different problem types.
Possible differences between problem types and solution types were addressed. Results showed that the difference between insight and analytical problems is significant only for one scale: insight problems are evaluated as more sudden in comparison to analytical problems. A comparison between solution types (insightful or step-by-step) gave significant results in three subjective scales: insightful solutions score higher in pleasure, suddenness and relief. The Chi-square test showed that the problem type and the solution type are related to each other: insight problems are more likely to be solved with insight, and analytical problems are more likely to be solved by step-by-step way. At the same time, there is a large number of solu-
a Insightful solution • Step-by-step solution
* 1
Pleasure Surprise Suddenness Relief Certainty Drive
Table 2
Contingency Tables for the Problem Type and the Solution Type
Problem type Solution type
Insightful solution Step-by-step solution
Analytical problem Count 259.0 1064
Expected count 599.4 723.6
% within row 19.6% 80.4%
Insight problem Count 948.0 393.0
Expected count 607.6 733.4
% within row 70.7% 29.3%
tions where insight problems are solved step-by-step, and analytical problems are accompanied by an Aha! experience.
The findings support the previous empirical evidence that insight problems can be solved without insight (Weisberg, 2015; Danek et al., 2016; Webb et al., 2018), and analytical problems, on the contrary, can be solved with an Aha! experience (Webb et al., 2018; Lazareva & Vladimirov, 2019). However, our data contradict the Representational Change Theory (RCT). According to RCT, the Aha! experience is a consequence of an impasse, and a representational change is needed to break the impasse. At the same time, a representational change is perceived as an attribute of insight problems, but not — with rare exceptions — of analytical problems. Since in our sample almost 30% of classical insight problems were solved using a step-by-step strategy, and almost 20% of analytical problems were solved with an insight, we cannot define this situation as rare or accidental.
How can we interpret the obtained results? The first and most obvious interpretation is that RCT is mistaken in its predictions. The representational change is not related to the problem type and can occur much more frequently in different problem types than it was originally thought.
The second interpretation is that the discrepancy between the problem type and the solution type is related to the term of Aha! experience. As we wrote earlier, at the moment the relationship between the Aha! experience and the representational change is an open question (Ammalainen & Moroshkina, 2021; Becker et al., 2020; Becker et al., 2021; Cranford & Moss, 2012; Danek et al., 2016; Danek et al., 2020). It is unclear whether the Aha! experience occurs in response to a representational change or whether it reflects some other process: finding the correct solution unexpectedly quickly (Dubey et al., 2021); an unexpected result or a path of finding it (Savinova & Korovkin, 2022); changes in information processing fluency (Ammalainen & Moroshkina, 2021; Moroshkina et al., 2024). Another question also remains unanswered: is the presence of both the Aha! experience and a representational change a necessary condition for recognizing a solution as insightful, or is one attribute necessary and sufficient for this purpose?
The third interpretation is that existing contradictions are associated with the questionnaire of the Aha! experience. Firstly, the Aha! experience is understood as
a multi-dimensional construct that consists of different scales in Danek and Wiley's questionnaire (2017). According to the nonspecific approach to metacog-nitive processes, a multi-dimensional division is unnecessary, because metacogni-tive feelings perform a signaling function reporting the result of unconscious processes rather than the specific source and content of information (Tikhonov et al., 2018). Probably, the variety of metacognitive feelings within the Aha! experience arises from the influence of the measurement procedure: asking to rate the Aha! experience on six different scales forces participants to look for and attribute feelings that they did not initially have. Secondly, the questionnaire can give too high a score to analytic problems, preventing us from finding more differences between problem types and solution types. If we look closely at those scales on which insightful and step-by-step solutions do not differ, we will see that there are some problems with their logic or interpretation:
• Surprise: The moment of solution was... (not surprising — surprising). The scale can be interpreted in various ways. On the one hand, it can be understood as intended by the authors: "I feel surprised that I have understood something" (Danek & Wiley, 2017, p. 4). On the other hand, the scale has other interpretations: "I found the answer at unexpected moment", "I found the answer by accident", "I don't know how I was able to find the answer, the path to it is unknown to me". Depending on the interpretation of the scale, participants will give different answers, which ultimately reduces power of scale.
• Certainty: How certain you are that your solution is correct (uncertain — certain). The scale is understood unambiguously, but participants can be certain in both solution types, insightful and step-by-step. It is easy to imagine a situation where, after solving a simple analytical problem (e.g., 2 + 7 + 4 = 13), the participant reports that they are certain in the solution.
• Drive: I am looking forward to the next problem... (no — yes). The inclusion of the scale in the questionnaire of the Aha! experience is due to the fact that there should be an "energizing effect on problem solving behavior" (Ohlsson, 1984, p. 70). However, nothing says that the energetic effect should arise precisely from insightful solution, and not from finding the correct answer.
It may be worth taking a critical look at some scales of the questionnaire and conducting psychometric testing not only on the material of magic tricks, but also on classical insight problems. Besides, there was an attempt in this direction (Shumilov et al., 2023).
Finally, we would like to answer the question of how appropriate the term "insight problem" really is. Despite all of the above, we can conclude that relying only on the problem type is not the best strategy for investigation of insight, because insight and analytical problems significantly differ from each other only in one subjective scale of the Aha! experience. The use of solution types is much more fruitful and allows better distinguishing different situations in the process of problem solving. We recommend that subjective questionnaires are applied to determine that a solution has been achieved through insight, i.e., with an Aha! experience.
Limitations
Our joint analysis has limitations. First, not all interesting parameters were explicitly and clearly represented in the data. For example, not all studies used all scales of the questionnaire, and not all of them measured the solution time or noted the solution type (insightful or step-by-step). Second, some of the problem types included in this joint analysis were represented by only one experiment (for example, CRA problems) or were not present at all (for example, anagrams), since the multidimensional questionnaire does not contain them in the most studies. Thereby, some results may be distorted, because they are only related to those problems that we were able to find, rather than to the problem type as a whole. Thirdly, although we used various experimental data in the joint analysis, we could not control the diversity of such parameters as gender, age, profession, etc. Finally, our results show that there is an interaction between the problem type and the solution type, which could also influence the final results because these characteristics were not completely independent of each other. All these factors constitute significant limitations in the selection and data analyses, which narrows down the explanatory possibilities of this work.
Conclusions
Our joint analysis allows us to formulate the following conclusions and future directions:
1. Using solution types provides more benefits for investigation of insight than using problem types. Insightful solutions are more pleasant, sudden and relieving compared to step-by-step solutions.
2. It is necessary to explore the connection between a representational change and an Aha! experience more carefully. At present it remains unclear whether a representational change may be the cause of an Aha! experience and how essential both attributes are for recognizing a solution as insightful.
3. It is worth paying more attention to subjective questionnaires of the Aha! experience, since the accuracy of insight studies significantly depends on the accuracy of measurements. We propose to take a critical look at some scales of Danek and Wiley's questionnaire and to conduct psychometric testing not only on the material of magic tricks, but also on classical insight problems.
References
Ammalainen, A., & Moroshhkina, N. (2021). The effect of true and false unreportable hints on anagram
problem solving, restructuring, and the Aha!-experience. Journal of Cognitive Psychology, 33(6-7),
644-658. https://doi.org/10.1080/20445911.2020.1844722 Becker, M., Kühn, S., & Sommer, T. (2021). Verbal insight revisited — dissociable neurocognitive processes
underlying solutions accompanied by an AHA! experience with and without prior restructuring.
Journal of Cognitive Psychology, 33(6-7), 659-684. https://doi.org/10.1080/20445911.2020.1819297
A.ß. CaeuHoea u dp. MHcaümHaa 3adaua uäu UHcaümHoe peweHue
421
Becker, M., Wiedemann, G., & Kühn, S. (2020). Quantifying insightful problem solving: A modified compound remote associates paradigm using lexical priming to parametrically modulate different sources of task difficulty. Psychological Research, 84, 528-545. https://doi.org/10.1007/s00426-018-1042-3
Bilalic, M., Graf, M., Vaci, N., & Danek, A. H. (2021). The temporal dynamics of insight problem solv-ing-restructuring might not always be sudden. Thinking & Reasoning, 27(1), 1-37. https://doi.org/10.1080/13546783.2019.1705912 Bowden, E. M., Jung-Beeman, M., Fleck, J., & Kounios, J. (2005). New approaches to demystifying
insight. Trends in Cognitive Sciences, 9(7), 322-328. https://doi.org/10.1016/j.tics.2005.05.012 Chuderski, A., Jastrz^bski, J., & Kucwaj, H. (2021). How physical interaction with insight problems affects solution rates, hint use, and cognitive load. British Journal of Psychology, 112(1), 120-143. https://doi.org/10.1111/bjop.12442 Chuderski, A., Jastrz^bski, J., Kroczek, B., Kucwaj, H., & Ociepka, M. (2021). Metacognitive experience on Raven's matrices versus insight problems. Metacognition and Learning, 16(1), 15-35. https://doi.org/10.1007/s11409-020-09236-6 Cranford, E. A., & Moss, J. (2012). Is insight always the same? A protocol analysis of insight in compound remote associate problems. The Journal of Problem Solving, 4(2), Article 8. https://doi.org/10.7771/1932-6246.1129 Danek, A. H., Fraps, T., von Müller, A., Grothe, B., & Öllinger, M. (2014). It's a kind of magic — what self-reports can reveal about the phenomenology of insight problem solving. Frontiers in Psychology, 5. https://doi.org/10.3389/fpsyg.2014.01408 Danek, A. H., & Wiley, J. (2017). What about false insights? Deconstructing the Aha! experience along its multiple dimensions for correct and incorrect solutions separately. Frontiers in Psychology, 7. https://doi.org/10.3389/fpsyg.2016.02077 Danek, A. H., & Wiley, J. (2020). What causes the insight memory advantage? Cognition, 205, Article
104411. https://doi.org/10.1016/j.cognition.2020.104411 Danek, A. H., Wiley, J., & Öllinger, M. (2016). Solving classical insight problems without Aha! experience: 9 Dot, 8 Coin, and Matchstick Arithmetic Problems. The Journal of Problem Solving, 9(1), Article 4. https://doi.org/10.7771/1932-6246.1183 Danek, A. H., Williams, J., & Wiley, J. (2020). Closing the gap: connecting sudden representational change to the subjective Aha! experience in insightful problem solving. Psychological Research, 84(1), 111-119. https://doi.org/10.1007/s00426-018-0977-8 Drqzyk, D., Kumka, M., Zarzycka, K., Zguda, P., & Chuderski, A. (2020). No indication that the ego depletion manipulation can affect insight: a comment on DeCaro and Van Stockum (2018). Thinking & Reasoning, 26(3), 414-446. https://doi.org/10.1080/13546783.2019.1649191 Dubey, R., Ho, M. K., Mehta, H., & Griffiths, T. (2021). Aha! moments correspond to meta-cognitive
prediction errors. https://doi.org/10.31234/osf.io/c5v42 Jung-Beeman, M., Bowden, E. M., Haberman, J., Frymiare, J. L., Arambel-Liu, S., Greenblatt, R., Reber, P. J., & Kounios, J. (2004). Neural activity when people solve verbal problems with insight. PLoS Biology, 2(4), Article e97. https://doi.org/10.1371/journal.pbio.0020097 Korovkin, S., Savinova, A., Padalka, J., & Zhelezova, A. (2021). Beautiful mind: grouping of actions into mental schemes leads to a full insight Aha! experience. Journal of Cognitive Psychology, 33(6-7), 62—630. https://doi.org/10.1080/20445911.2020.1847124 Lazareva, N. Yu., & Vladimirov, I. Yu. (2019). The influence of fixedness on the formation problem incorrect representation and the emergence of insight solutions. Uchenye Zapiski Rossiiskogo
Gosudarstvennogo Sotsial'nogo Universiteta [Scientific Notes of Russian State Social University], 18(4(153)), 22-30. https://doi.org/10.17922/2071-5323-2019-18-4-22-30 (in Russian) Moroshkina, N. V., Pavliuchik, E. I., Ammalainen, A. V., Gershkovich, V. A., & Lvova, O. V. (2024). The Aha! experience is associated with a drop in the perceived difficulty of the problem. Frontiers in Psychology, 15, Article e1314531. https://doi.org/10.3389/fpsyg.2024.1314531 Novick, L. R., & Sherman, S. J. (2003). On the nature of insight solutions: Evidence from skill differences in anagram solution. The Quarterly Journal of Experimental Psychology Section A, 56(2), 351382. https://doi.org/10.1080/02724980244000288 Ohlsson, S. (1984). Restructuring revisited: I. Summary and critique of the Gestalt theory of problem solving.
ScandinavianJournal of Psychology, 25(1), 65-78. https://doi.org/10.1111/j.1467-9450.1984.tb01001.x Ohlsson, S. (1992). Information-processing explanations of insight and related phenomena. In K. J. Gilhooley (Ed.), Advances in the psychology of thinking (pp. 1-44). London: Harvester-Wheatsheaf.
Ohlsson, S. (2011). Deep learning: How the mind overrides experience. Cambridge University Press. Salmon Mordekovich, N., & Leikin, M. (2022). The cognitive-creative profiles of insightful problem solvers: A person centered insight study. The Journal of Creative Behavior, 56(3), 396-413. https://doi.org/10.1002/jocb.536 Savinova, A., & Korovkin, S. (2022). Surprise! Why insightful solution is pleasurable. Journal of
Intelligence, 10(4), Article 98. https://doi.org/10.3390/jintelligence10040098 Shumilov, T. V., Chistopolskaya, A. V., & Vladimirov, I. Y. (2023). The path to insight: Developing methods to increase the effectiveness of insightful solution detection. Psychology. Journal of the Higher School of Economics, 20(3), 473-489. https://doi.org/10.17323/1813-8918-2023-3-473-489 Spiridonov, V., Loginov, N., & Ardislamov, V. (2021). Dissociation between the subjective experience of insight and performance in the CRA paradigm. Journal of Cognitive Psychology, 33(6-7), 685699. https://doi.org/10.1080/20445911.2021.1900198 Tikhonov, R. V., Ammalainen, A. V., & Moroshkina, N. V. (2018). The variety of metacognitive feelings: Different phenomena or different terms? Vestnik Sankt-Peterburgskogo Universiteta. Seriya "Psikhologiya i Pedagofika" [Vestnik of Saint Petersburg University. Psychology and Education], 8(3), 214-242. https://doi.org/10.21638/11701/spbu16.2018.302 (in Russian) Vladimirov, I. Yu., Bushmanova, A. S., & Makarov, I. N. (2021). Behavioral markers of key events of an insight solution. Psikhologicheskie Issledovaniya [Psychological Studies], 14(77). https://doi.org/10.54359/ps.v14i77.134 (in Russian) Webb, M. E., Little, D. R., & Cropper, S. J. (2016). Insight is not in the problem: Investigating insight in problem solving across task types. Frontiers in Psychology, 7. https://doi.org/10.3389/fpsyg.2016.01424 Webb, M. E., Little, D. R., & Cropper, S. J. (2018). Once more with feeling: Normative data for the aha experience in insight and noninsight problems. Behavior Research Methods, 50(5), 2035-2056. https://doi.org/10.3758/s13428-017-0972-9 Weisberg, R. W. (1995). Prolegomena to theories of insight in problem solving: A taxonomy of problems. In R. J. Sternberg & J. E. Davidson (Eds.), The nature of insight (pp. 157-196). The MIT Press.
Weisberg, R. W. (2015). Toward an integrated theory of insight in problem solving. Thinking &
Reasoning, 21(1), 5-39. https://doi.org/10.1080/13546783.2014.886625 Wong, T. J. (2009). Capturing 'Aha!' moments of puzzle problems using pupillary responses and blinks [Doctoral dissertation]. University of Pittsburgh, PA, USA. https://core.ac.uk/download/pdf/12208737.pdf
