“Aha!” is right: Insight solutions are more likely to be correct than are analytic solutions Carola Salvi1, Azurii K. Collier2, Emanuela Bricolo1, John Kounios3, Mark Beeman2 1Dept. of Psychology, Milano-Bicocca University; 2Dept. of Psychology, Northwestern University, Evanston; 2 Dept. of Psychology, Drexel University c.salvi4@campus.unimib.it; mjungbee@northwestern.edu Archimedes was taking a bath when the solution of how to assess the volume We analyzed all the problems participants claimed they solved, both of an irregular object (the King’s crown), suddenly arose in his mind. problems answered correctly and incorrectly. Astonished, he jumped from the tub and ran through the streets screaming "Eureka!” Results: Conclusions: Solution % of problems Archimedes could have solved his problem via: Preliminary experiment type solved (sd) N = 40 correct insight 91 ( +_ 8) Insight responses are more likely to Insight Analysis 89% analysis 81 ( +_ 21) 90 CRA Time limit: 12 seconds incorrect insight 9 ( +_ 8) be correct than analysis responses • Intermediate solving steps unknown OR •Intermediate steps or processes known 11% analysis 19 ( +_ 21) 51 % of problems were claimed to be solved, 60% of • Solution arises suddenly as a whole • Solution gradually arises which via insight Correct t (39) = 2.48, p < .01 2x2 ANOVA interaction (F (1,39)= 9.73, p < .005 Insight depends on the integration of • Processes leading to solution • Processes or strategy reportable multiple weak associations from all three unreportable • Solution unsurprising, but needs Experiment Problems claimed to problem words to solution; summation N = 21 insight correct incorrect • Solution feels obvious, surprising testing 120 CRA be solved boosts activation of the solution concept 45% 58% 88% 12% How did Archimedes’ insight emerge? Time limit: 15 seconds into consciousness. This integration of problem concepts Initial in-depth analysis Weak activation of remote solution-related Is non-obvious for the individual, and As a consequence, when insight solution info; remains outside of awareness therefore contextually non-biased [2] emerges, it necessarily fits all three Through convergence & integration, solution Is modulated by the activation of anterior words emerges as a whole, tied to all problem cingulate cortex (ACC) [5] elements ACC activation[6] suggests changes in top- Switch from unconscious to conscious • Detection of competing responses down cognitive control. In insight: • Shift attention to alternative ideas processing, triggers the sudden ‘aha!’ ACC is also related to: • Does not suppress weak (but related) feeling [3,4] response competition and error detection info Was Archimedes’ Eureka just “lucky” or was his public display of • Discriminates the relevant info from enthusiasm actually supported by a higher likelihood of being correct? Correct t (20) = 3.37, p < .005 noise Incorrect t (18) = -2.43, p < .05 2x2 ANOVA interaction (F (1,18) = 15.25, p < .005 • Allows summation/integration Method & Procedure • Enables detection of weak info, to Task switch attention and recognize solution Problems solved in less than 2 and more than 10 seconds were removed as immediate Compound Remote Associate problems (CRA): Three words, each forming a recognitions and last attempts to guess, respectively. compound word or phrase with solution (e.g., pine/crab/sauce —APPLE). Alternative interpretation: Solution to each problem obtained either via insight or via analysis [1, 3, 6] Perhaps there is lower probability to make an error in a one-step insight process, + Problems claimed to compared to multi-step analytic process, be solved insight correct incorrect 35% 63% 92% 8% where each step is prone to compounded Ready? error. 1 sec + References: 1. Bowden, E. M., & Jung-Beeman, M. (2003). Normative data for 144 compound remote associate problems. Behavior research methods, instruments, & computers: a journal of the Psychonomic Society, Inc, 35(4), sec? pine 634-9. crab 2. Bowden, E. M., Jung-Beeman, M., Fleck, J., & Kounios, J. (2005). New approaches to demystifying insight. sauce Trends in cognitive sciences, 9(7), 322-8. 3. Jung-Beeman, M., Bowden, E. M., Haberman, J., Frymiare, J. L., Arambel-Liu, S., Greenblatt, R., Reber, P. Tell the solution to the experimenter J., et al. (2004). Neural activity when people solve verbal problems with insight. PLoS biology, 2(4), E97. 1 sec Press left button for insight Correct t (20) = 3.71, p < .005 4. Kounios, J., Beeman, M. (2009). Aha!Moment: The Cognitive Neuroscience of Insight. Current Directions in Psychological Science, 21(4), 415-216. Press right button for analytic 2x2 ANOVA interaction (F (1,18) = 14.74, p < .005 5. Kounios, J., Frymiare, J. L., Bowden, E. M., Fleck, J. I., Subramaniam, K., Parrish, T. B., & Jung-beeman, M. (2006). Subsequent Solution by Sudden Insight. Psychological Science, 17(10), 882-890. 12-15 sec max 6. Subramaniam, K., Kounios, J., Parrish, T. B., & Jung-Beeman, M. (2009). A brain mechanism for facilitation of insight by positive affect. Journal of cognitive neuroscience, 21(3), 415-32.