Short CommunicationThe dark side of Eureka: Artificially induced Aha moments make facts feel true
Introduction
John Nash, a mathematician and Nobel laureate, was asked why he believed that he was being recruited by aliens to save the world. He responded, “…the ideas I had about supernatural beings came to me the same way that my mathematical ideas did. So I took them seriously” (Nasar, 1998). Although Nash was diagnosed with Schizophrenia in 1959, the example exposes a basic human conundrum. In everyday life humans need to discern the difference between a true and useful idea and a false one, and sometimes must do so quickly in order to respond in conversation, give advice, or solve a problem under pressure. How is the validity of a new idea evaluated, especially when time is of the essence? Perhaps the metacognitive process described by Nash is correct, and humans turn to the phenomenology that accompanies their ideas—their Aha! moments.
Ideas that are called ‘insights’ are defined by metacognitive suddenness (Metcalfe & Wiebe, 1987) and an immediate sense that the idea is correct or valuable despite its unexpected appearance in mind (Ohlsson, 1984; Kounios & Beeman, 2014; Danek & Wiley, 2017; Laukkonen & Tangen, 2017). Recent empirical work suggests that when participants report an Aha! experience—the subjective marker of insight—then the solution they provide tends to be correct (Danek, Fraps, von Müller, Grothe, & Öllinger, 2014; Hedne, Norman, & Metcalfe, 2016; Salvi, Bricolo, Kounios, Bowden, & Beeman, 2016; Webb, Little, & Cropper, 2016; Danek & Wiley, 2017). For example, Salvi et al. (2016) presented participants with four different problems to solve: compound remote associates, anagrams, rebus puzzles, and degraded images. For each of the problems, when participants reported a feeling of Aha! they were more likely to provide a correct response (nearly twice as likely in some cases). This insight-accuracy effect appears to be robust across a number of laboratory problems, and effect sizes are consistently large (e.g., Hedne et al., 2016; Webb et al., 2016; Danek & Wiley, 2017).
There is currently no generally accepted explanation for why the feeling of insight should predict accurate solutions to problems, but there are theoretical frameworks for which it is not so surprising. According to Feelings-as-Information Theory (Schwarz, 2011), subjective experiences in the forms of emotions, bodily sensations, and metacognitive experiences are sources of information that humans regularly rely on to make judgments and decisions (see also Bechara, Damasio, & Damasio, 2000; Slovic, Finucane, Peters, & MacGregor, 2007). Obvious examples include hunger, fear, pleasure, and tiredness, which signal something about the organism’s internal state, or an automatic appraisal of some external phenomenon. The role of feeling in guiding decision-making has been demonstrated in far-reaching domains including risk judgments (Fischhoff, Slovic, Lichtenstein, Read, & Combs, 1978), stock market investments (Alter & Oppenheimer, 2006; Hirshleifer & Shumway, 2003), gambling and probability judgments (Loewenstein, Weber, Hsee, & Welch, 2001), truth and memory judgments (Reber & Schwarz, 1999; Dougal & Schooler, 2007; Schwarz, Sanna, Skurnik, & Yoon, 2007), and jury decision-making (Semmler & Brewer, 2002). It is feasible that the Aha! experience—like the many other feelings and sensations that guide decision-making in productive ways—is a source of information for the problem solver. Moreover, if the feeling of insight carries information about the veracity of a new solution—as it subjectively purports to—then it would not be surprising that it predicts accurate solutions.
How might Aha! moments carry information about the veracity of a new idea? When a scientist, an inventor, or an artist has a new idea, they may be drawing on a vast repository of knowledge and expertise. Therefore, one possibility is that the insight experience signals that the new idea is highly coherent with the individual’s existing knowledge and experiences. It’s well known that experts can automatically and intuitively bring their expertise to bear in their domain, often without explicitly knowing why their intuitions are correct (Ericsson & Charness, 1994; Kahneman, 2015). New ideas may be evaluated through a similar mechanism, the only difference being that the idea occurs ‘in the head’ for the problem solver, whereas the stimulus is ‘in the world’ for the expert. Therefore, when a solution to a problem appears in mind, the problem solver can use the Aha! experience as a heuristic shortcut—a quick appraisal of whether the idea is consistent with what they know—instead of engaging in a slow and effortful evaluation (Slovic et al., 2007; Gigerenzer & Gaissmaier, 2011). So long as the person’s existing knowledge is valid, then the Aha! experience will likely signal a correct solution. For a detailed theoretical discussion of our theory—termed the Eureka Heuristic—see Laukkonen, Schooler, and Tangen (2018), and Laukkonen (2019).
This experiment is based on a specific prediction that arises from the view described above. If humans are being guided by their Aha! experiences as signals of veracity, then presumably artificially induced Aha! phenomenology ought to bias judgments. Similar effects have been found using feelings of surprise (Whittlesea & Williams, 2001), fluency (Reber & Schwarz, 1999, addressed further in the discussion), and familiarity (Whittlesea, Jacoby, & Girard, 1990). To test this prediction, we borrowed the “discovery misattribution” paradigm used by Dougal and Schooler (2007) in which participants confused the experience of successfully deciphering scrambled words with that of having previously studied them. Here we presented participants with claims such as: ‘ithlium is the lightest of all metals’, where the scrambled word is ‘lithium’ (Reber & Unkelbach, 2010). Participants needed to solve the scrambled word before the proposition is complete, and then they rated the extent to which they believe that the proposition is true. We expected that successfully solving the anagram will induce an Aha! experience that would be misattributed to the proposition, leading to biased truth judgments.
Our main interest was comparing truth judgments within-participants for solved and unsolved anagrams (with and without Aha!). However, we also included a between-subjects variable so that we could investigate baseline truth judgments without the presence of an anagram. If we find a baseline difference between the presence of the anagram and no anagram, this is equivalent to finding the ‘Revelation Effect’ (see Watkins & Peynircioglu, 1990; Bernstein, Whittlesea, & Loftus, 2002). We also included a condition where the key word—the same word that was scrambled in the anagram condition—was presented after a short delay. Solving an anagram inevitably leads to a delayed presentation of the key word that completes the proposition, and we wanted to ensure that the delayed presentation (which may itself elicit a miniature Aha! moment) was not accounting for any effects we observe. In the interest of a brief report, we provide detailed hypotheses, prespecified decision rules, instruction transcripts, and exploratory analyses on the Open Science Framework (OSF): https://osf.io/up98z.
Section snippets
Design & materials
This experiment was approved by the University of California, Santa Barbara, Human Subjects Committee, clearance number: 81-18-0543, in accordance with the Declaration of Helsinki. The experiment had three within subject variables: 2 (Proposition: true or false) × 2 (Problem: solved or unsolved) × 2 (Aha! Experience: yes or no), and one between subjects factor (Anagram: present, absent, and absent with delay). The dependent measure was truth judgments on a 12 point scale ranging from 1
Descriptives
After applying our decision rules, 268 of the 300 participants were included in the analyses. On average, participants solved the anagrams 59.6 % of the time (SD = .3), and the mean accuracy for individual anagrams is shown in Fig. 2. Unsurprisingly, participants provided higher truth ratings for true claims (M = 6.92, SD = 1.39), and lower ratings for false claims (M = 5.9, SD = 1.44), and the difference was significant, t(267) = 13.8, p < .001, d = .84. The anagrams elicited insights 39 % of the time,
Discussion
There is a certain mystery about an idea that suddenly strikes the conscious mind, as if totally complete and true. The past century of research has progressed our understanding of the kinds of problem-solving processes that precede sudden solutions, and the best way to elicit insight experiences (Maier, 1931; Ohlsson, 1984; Schooler & Melcher, 1995; Sternberg & Davidson, 1995 Öllinger & Knoblich, 2009; Ohlsson, 2011; Laukkonen & Tangen, 2018). A less explored level of analysis is the role of
Acknowledgments
This research was supported by grant no. 44069-59380 from the Fetzer Franklin Fund to J.W.S.
References (46)
- et al.
Working Wonders? Investigating insight with magic tricks
Cognition
(2014) - et al.
The effect of nonprobative photographs on truthiness persists over time
Acta Psychologica
(2013) - et al.
Can observing a Necker cube make you more insightful?
Consciousness and Cognition
(2017) - et al.
Enabling spontaneous analogy through heuristic change
Cognitive Psychology
(2017) - et al.
Effects of perceptual fluency on judgments of truth
Consciousness and Cognition
(1999) - et al.
Metacognitive experiences and the intricacies of setting people straight: Implications for debiasing and public information campaigns
Advances in Experimental Social Psychology
(2007) - et al.
The affect heuristic
European Journal of Operational Research
(2007) - et al.
Availability: A heuristic for judging frequency and probability
Cognitive Psychology
(1973) - et al.
Illusions of immediate memory: Evidence of an attributional basis for feelings of familiarity and perceptual quality
Journal of Memory and Language
(1990) - et al.
Predicting short-term stock fluctuations by using processing fluency
Proceedings of the National Academy of Science
(2006)
Emotion, decision making and the orbitofrontal cortex
Cerebral Cortex
Increasing confidence in remote autobiographical memory and general knowledge: Extensions of the revelation effect
Memory and Cognition
REBUS and the anarchic brain: toward a unified model of the brain action of psychedelics
Pharmacological Reviews
What about false insights? Deconstructing the Aha! experience along its multiple dimensions for correct and incorrect solutions separately
Frontiers in Psychology
Discovery misattribution: When solving is confused with remembering
Journal of Experimental Psychology: General
Expert performance: Its structure and acquisition
The American Psychologist
How safe is safe enough? A psychometric study of attitudes towards technological risks and benefits
Policy Sciences
Heuristic decision making
Annual Review of Psychology
Intuitive feelings of warmth and confidence in insight and noninsight problem solving of magic tricks
Frontiers in Psychology
Good day sunshine: Stock returns and the weather
The Journal of Finance
Thinking, fast and slow
Subjective probability: A judgment of representativeness
The concept of probability in psychological experiments
The cognitive neuroscience of insight
Annual Review of Psychology
Cited by (42)
Insight and the selection of ideas
2023, Neuroscience and Biobehavioral ReviewsRestructuring insight: An integrative review of insight in problem-solving, meditation, psychotherapy, delusions and psychedelics
2023, Consciousness and CognitionThinking style and psychosis proneness do not predict false insights
2022, Consciousness and CognitionCitation Excerpt :The researchers found that when the pictorial hint was relevant to the shorter incorrect solution, participants were more likely to arrive at that solution than when the hint was associated with the correct solution. Insight phenomenology, even for correct solutions, can also be misleading when the insight experience itself triggers false beliefs through a process of “misattribution” (Laukkonen et al., 2021; Laukkonen et al., 2020). For example, if a proposition such as “humans have free will” is presented at the same time that an ‘Aha’ moment is artificially elicited, participants will tend to rate the proposition as more believable or true (Laukkonen et al., 2021).
A Wandering Mind is Not Always a Creative Mind: How Thought Dynamics Explain the Relationship between Mind Wandering and Creativity
2024, Journal of Creative Behavior