Ben Gold, a PhD candidate Lab of Robert Zatorre The Neuro (Montreal Neurological Institute and Hospital) McGill University

Music Activates Central Hub of Brain’s Reward System Interview with:

Ben Gold, a PhD candidate Lab of Robert Zatorre The Neuro (Montreal Neurological Institute and Hospital) McGill University

Ben Gold

Ben Gold, a PhD candidate
Lab of Robert Zatorre
The Neuro (Montreal Neurological Institute and Hospital)
McGill University What is the background for this study? What are the main findings? 

Response: Music is just sound in air, but it carries considerable power. It captivates our brain’s reward system, we devote an enormous amount of time and money to it, and we’re just beginning to tap its therapeutic potential. We wanted to explore how something so abstract could have such an impact, and since music is so well suited to establishing and manipulating patterns of sound as it unfolds, we focused on how it manipulates expectations.

Previous research has shown that surprises are often the most emotional and pleasurable moments in music listening, but whether and how this engaged the brain’s reward system was unclear. So we adapted an experimental protocol designed for studying learning and surprise about more concrete rewards like food or money, and applied it to a musical context during brain imaging. This protocol relies on participants making decisions from which we could infer their expectations, allowing us to estimate how surprised they were by each outcome whenever it occurred. In our case, we asked participants to make choices about colors and directions that were associated with different musical outcomes, but we didn’t tell them what those associations were so that they they started with no expectations and learned as they went.

We found that our participants could learn about music just like they would learn about how to find food or win money, and that the same neural process was involved. Specifically, we saw that the activity of the nucleus accumbens — a central hub of the reward system — reflected both how pleasant and how surprising each musical outcome was: a computation known as a reward prediction error. Across individuals, those who better represented these reward prediction errors in their nucleus accumbens also learned better about the music in the experiment, making more decisions over time to find the music they preferred. What should readers take away from your report?

Response: Our study is the first to our knowledge showing that reward prediction errors — well known for their role in learning about adaptive rewards like food, money, erotic stimuli, etc. — could also be involved in aesthetic pursuits like the enjoyment of music. While music is especially good at manipulating expectations, we suspect that many other forms of entertainment capitalize on the same process. That the reward system gets involved in learning, even about something with no apparent biological value, also emphasizes the ability of surprise to motivate and captivate us. We therefore hope that this work leads to a richer understanding of the processes underlying curiosity and learning, as well as the cognitive contributions to emotions like the pleasure of music listening. What recommendations do you have for future research as a result of this work?

Response: While this study provides the first direct evidence that we know of for reward prediction errors in an aesthetic context, we would like to see more research expanding our findings to other domains and to a better understanding of the mechanism through which abstract stimuli elicit and manipulate expectations. We would also like to see research that looks at more naturalistic experiences of music listening, to test whether the same processes are involved in passive, social, and other forms of typical music listening.

Finally, the exploration of reward-related surprise begs the question of what makes a surprise pleasant or unpleasant: Is it the timing? The amount of surprise? The listener’s experience? Is there anything else you would like to add?

Response: I am grateful to my co-authors, participants, and research assistants for their invaluable contributions to this publication. I also thank Fulbright Canada for their STEM Graduate Award, the Natural Sciences and Engineering Research Council-CREATE program for their fellowship, the Canadian Institutes of Health Research for their foundation grant, and the Canadian Institute for Advanced Research for their senior fellowship that funded this research.

Citation:Benjamin P. Gold, Ernest Mas-Herrero, Yashar Zeighami, Mitchel Benovoy, Alain Dagher, Robert J. Zatorre. Musical reward prediction errors engage the nucleus accumbens and motivate learning. Proceedings of the National Academy of Sciences, 2019; 201809855 DOI: 1073/pnas.1809855116

[wysija_form id=”3″]


The information on is provided for educational purposes only, and is in no way intended to diagnose, cure, or treat any medical or other condition. Always seek the advice of your physician or other qualified health and ask your doctor any questions you may have regarding a medical condition. In addition to all other limitations and disclaimers in this agreement, service provider and its third party providers disclaim any liability or loss in connection with the content provided on this website.


Last Updated on February 20, 2019 by Marie Benz MD FAAD