Decoding episodic autobiographical memory in naturalistic virtual reality.

Area of Science:

• Cognitive Neuroscience
• Psychology

Background:

• Episodic autobiographical memory (EAM) involves recalling personal events with context and subjective elements.
• EAM formation lacks controlled, real-life-like study paradigms and predictive models for long-term retrieval based on encoding subjectivity.

Purpose of the Study:

• To investigate incidental EAM encoding in an immersive virtual environment.
• To develop a predictive model for long-term EAM retrieval using subjective encoding experiences.
• To analyze the temporal dynamics of EAM predictors.

Main Methods:

• A longitudinal study with 30 participants using an immersive virtual environment.
• Incidental encoding of events with varying emotional valences, involving interaction or observation.
• Three surprise free recall tests at immediate, one-week, and one-month intervals post-encoding.
• Simultaneous physiological recordings during event encoding.

Main Results:

• Predictive analyses revealed temporal dynamics in EAM predictors derived from subjective encoding ratings.
• Sense of remembering and event infrequency were consistent predictors over time.
• Emotion, mental imagery, and prospective aspects emerged as time-dependent predictors.
• This highlights a dynamic, time-dependent role for memory predictors, challenging uniform influence theories.

Conclusions:

• EAM retrieval predictors exhibit dynamic, time-dependent characteristics.
• Memory formation appears multiphasic, involving different mechanisms during encoding, consolidation, and retrieval.
• This challenges traditional views of uniform encoding factor influence on long-term memory.