Graduation Year


Date of Submission


Document Type

Open Access Senior Thesis


Best Senior Thesis in Neuroscience

Degree Name

Bachelor of Arts



Reader 1

Alison Harris

Reader 2

Cathy Reed


Research in economics and neuroscience has shown that an item’s value is subjective, in that it depends on the circumstances and preferences of the observer rather than the item’s inherent properties. In particular, value can be affected by the context in which an item is presented: for example, an apple might look more appealing if it is offered right after something less tasty, such as broccoli, than after something tempting like chocolate cake. Consistent with this idea, studies using invasive electrophysiology have shown that value signals corresponding to the current item are modulated by the value of the previously encountered item. However, these findings have largely been observed in animals using small sets of relatively simple, highly trained stimuli. In contrast, real-world human decision making often involves a broad range of complex stimuli presented in close succession, such as during grocery shopping or online purchasing. Therefore, it remains an important question whether the value of a previously encountered item can affect the value signal of the current stimulus in the human brain. Here, we examined this question by measuring the effect of previous offer value on event-related potentials (ERP) in hungry human participants choosing whether to eat a large variety of snack foods. Previous analysis of this dataset had identified neural signals correlated with subjective value emerging approximately 400-600 ms after stimulus onset in central sensors. Comparing trials varying in “value distance,” defined as the absolute difference in subjective value between the previous and current item, we found that ERP value signals were largest when the change in value was most extreme. This effect emerged from 530 ms to 800 ms post-stimulus onset and, critically, emerged in central sensors previously associated with neural value signals. This study expands on previous literature to show that modulation of value signals relative to the value of the previously encountered item extends into naturalistic decision-making tasks in healthy humans.

Available for download on Tuesday, May 02, 2023