Proceedings of the Annual Meeting of the Cognitive Science Society

Many studies have shown that observers can accuratelyperceive and evaluate the statistical summary of presentedobjects’ attribute values, such as the average, withoutattending to each object. However, it remains controversialhow the visual system integrates the attribute values (e.g.,information on size) of multiple items and computes theaverage value. In this study, we tested how distributions ofitem sizes affect the precision and bias in judging averagevalues. We predicted that if observers utilize all of theavailable size information equally, the distribution wouldhave no effect, and vice versa. Our results showed that, withnovice observers, judgement precision differed among sizedistributions and that the observers overestimated the size ofthe average value compared to the actual size under allconditions. These results imply that observations of someitems in a set could be weighted more easily than others, withthe possibility that this process is easier for larger items thansmaller ones. However, this was not the case for experiencedobservers, who showed no effects of distribution type onaverage assessment performance. Our findings imply that theprocess of representing the average value may not beexplained by a single definitive mechanism and, is rathermediated by a mixture of multiple cognitive processes.