Sound expectations: Formation, maintenance and flexible adjustment of auditory predictive models

• Applicant: Dr. Nina Coy
• Subject Area: Biological Psychology and Cognitive Neuroscience
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 548511234

Modern predictive processing theories assume that prior information and previous experience, such as more or less regularly recurring auditory information, are used by the auditory system to build hierarchical internal models of the environment. These models facilitate expectations about which auditory events are likely to be encountered in the near future. In the laboratory, auditory predictive processing is typically studied by establishing some auditory regularity, and then comparing brain responses and/or behavior between those stimuli conforming to the established regularity (standards) and those violating it (deviants). While there is consensus that deviants are a powerful tool to elicit prediction error processing and/or probe the extraction of the standard regularity, there is no clear understanding yet of what happens to the underlying model post-deviation. In fact, the make-up of auditory internal models is generally not well understood. The present project will uncover the composition of internal models by systematically investigating the effect of deviants on and their potential representation in internal models. Shedding light on this novel issue - i.e., the role of deviants for shaping internal models - will be possible by specifically tailored adaptations of classical experimental paradigms (e.g., oddball, roving standard). In a series of experiments, the novel paradigms will be combined with behavioral and electroencephalographic measures (e.g., auditory event-related potentials) for characterizing the involved mental processes. With these methods, the project will first investigate how current behavioral goals modulate what information is represented in and selected from internal models, testing for flexible adaptation according to the current needs of the listener. Second, the project will examine the flexibility in forming and selecting auditory predictive representations, including the simultaneous representation of multiple internal models (i.e., auditory regularities) spanning multiple time scales. Third, the project will explore the function of deviants and their predictive potential in relation to contextual consequences, contrasting transient deviants (i.e. temporary divergence from the established regularity) with persistent deviations that turn into a new standard regularity. Altogether, the project aims to reveal the relevance and interplay of different kinds of predictive information, advance our conception of auditory internal models, and uncover how flexible the auditory system is in maintaining and adjusting these models according to current requirements. The findings from this project will not only be relevant to the field of auditory processing, but also to predictive processing and cognitive flexibility in general.

• DFG Programme: Research Grants
• Co-Investigators: Professorin Dr. Alexandra Bendixen; Dr. Sabine Grimm