Influence of Benchmarks on the Left Digit Effect in Number Line Estimation

Faculty Sponsor: Andrea L. Patalano

Prakriti Mittal
Prakriti Mittal

Prakriti Mittal is a rising senior (’23) from Mumbai, India. She graduated from Bombay International School. Her interests are swimming, playing with her dog, cooking and working at the museum on campus. She is majoring in Psychology and plans to go to graduate school after Wesleyan.

Abstract: The left digit effect is a numerical bias in which numbers with nearly identical magnitudes but different left digits are estimated to be significantly different from each other (eg., 699 and 701). It is also found in number line estimation tasks which are reliable predictors of math outcomes. Previous studies, using motivational incentives and accuracy feedback,have been unsuccessful in reducing the left digit effect. We tested the use of a benchmark intervention to reduce the left digit effect as it has been shown to reduce accuracy error generally. Participants (N = 165 adults) completed three blocks of 60 trials each of a 0-1000 number line task, where the second block had benchmarks for half of the participants. We found a statistically significant interaction, indicating a sustained reduction in the left digit effect during and following the second block with the benchmark intervention. There was also a significantly positive correlation between one’s left digit effect in the first block of the number line estimation task and one’s left digit effect one a more complex consumer judgment task.

pmittal_BCPJ_QAC_07222022