Choice in multi-alternative environments: a trial-by-trial implementation of the sequential choice model

Area of Science:

• Behavioral economics
• Animal behavior
• Comparative psychology

Background:

• The sequential choice model (SCM) predicts preferences based on individual option acceptance times.
• Previous SCM studies focused on binary choices (two alternatives).
• Testing SCM with more than two options is crucial for its generalizability.

Purpose of the Study:

• To test the predictive power of the sequential choice model (SCM) in a more complex choice environment.
• To assess SCM's ability to predict preferences among multiple alternatives in non-human animals.
• To validate SCM using European starlings (Sturnus vulgaris) with four background options.

Main Methods:

• European starlings were trained in an environment with four alternatives varying in delay to reinforcement.
• Binary choices between all six possible pairs of alternatives were presented.
• Response latencies and choice data were recorded to test SCM predictions.

Main Results:

• SCM predictions of preference strength showed good correspondence with observed choices across six choice situations.
• A trial-by-trial analysis revealed that SCM correctly predicted 84% of individual choices.
• The model demonstrated predictive validity in a complex choice scenario with multiple alternatives.

Conclusions:

• The sequential choice model (SCM) effectively predicts choice preferences even when multiple alternatives are available.
• SCM's applicability extends beyond simple binary choices, supporting its robustness in behavioral economics.
• This study validates SCM using a non-human animal model, highlighting its potential for broader application in understanding decision-making.