A test of affect processing bias in response to affect regulation

Area of Science:

• Neuroscience
• Cognitive Science
• Machine Learning

Background:

• Affect processing is influenced by internal states.
• Understanding neural mechanisms of affect regulation is crucial.

Purpose of the Study:

• To investigate how self-induced affect processing states bias subsequent emotional perception.
• To develop and validate a novel neuroimaging paradigm for studying affect regulation.

Main Methods:

• Combined machine learning with real-time functional magnetic resonance imaging (rtfMRI).
• Developed a paradigm where participants regulated brain states towards positive valence.
• Used neural decodings of rtfMRI to trigger affective image stimuli.
• Employed general linear mixed-effects models to analyze neural decoding of affect processing.

Main Results:

• Self-induction of a positive valence brain state significantly biased subsequent valence processing.
• Validated affect induction using psychophysiological measures (hedonic valence, autonomic arousal).
• Confirmed predictive accuracy of neural decoding models with out-of-sample data.

Conclusions:

• Novel experimental paradigms triggered by cognitive states are viable.
• Self-induced affective states play a significant role in biasing affect processing.
• This research offers a tool for exploring intrinsic cognitive mechanisms in perception.