Mice plan decision strategies based on previously learned time intervals, locations, and probabilities

Area of Science:

• Neuroscience
• Animal Behavior
• Cognitive Science

Background:

• Animals exhibit sophisticated timed behaviors influenced by learned probabilities.
• The cognitive mechanisms underlying these timed decisions, whether model-based or trial-and-error, remain unclear.

Purpose of the Study:

• To investigate whether mice utilize previously learned task parameters or gradual trial-and-error to adjust timed decisions.
• To characterize the emergence and temporal dynamics of switching behavior in mice.

Main Methods:

• Mice were trained on a timed-switching task with varying probabilities of short and long delay trials.
• Behavioral data was collected during a test phase where mice had to anticipate reward timing and location.
• Analysis focused on the emergence and stability of timed-switching behavior and adjustments based on trial probability.

Main Results:

• Mice displayed immediate and stable timed-switching behavior from the initial test trials.
• Behavioral adjustments, such as longer waits before switching, occurred in response to increased short-trial probability.
• Performance aligned with reward maximization predictions.

Conclusions:

• Mice abruptly adopted temporal decision strategies, directly integrating learned task parameters.
• Findings support a model-based representational account of temporal risk assessment in animals.
• This contrasts with a gradual, trial-and-error adjustment of time-dependent choice behavior.