choice
Definition
Choice refers to the allocation of behavior across available alternatives, defined operationally by the response units through which those alternatives are contacted. The identity of those units is not fixed: models have treated stays, switches, dwell times, bouts, and response sequences as the fundamental elements on which reinforcement contingencies operate, and these alternatives need not be mutually exclusive. Changeover delays (CODs) illustrate this point directly. When pigeons were trained on concurrent VI 30-s VI 60-s schedules with a COD assigned to one alternative, their responding organized into discrete bursts of rapid pecks, and peck-based preference for that alternative was more extreme than preference measured by dwell time, indicating that the observed preference reflected trained response structure rather than differential schedule value. Whether that structure transferred to unreinforced probe tests depended entirely on whether the probe procedure preserved the discriminative stimulus control that had been established during training.
Sources: Cleaveland (2026)
Related Terms
- response chunking (1 shared article)
- changeover delay (1 shared article)
- pigeons (1 shared article)
- variable-interval (1 shared article)
Applications
Choice and Changeover Delay
Changeover delays do more than reduce the probability that the first response to a new alternative will be reinforced. In concurrent VI VI schedules with pigeons, a 2.5-second COD reliably lengthened dwell times and organized responding into cohesive bursts of rapid pecks, and this response structure came under the discriminative control of the stimuli paired with the COD schedule. The resulting preference, as measured by response counts, exceeded that observed for a VI 30-s schedule without a COD, yet the dwell-time measures showed no comparable difference, demonstrating that CODs shape the behavioral units of choice independently of the underlying reinforcement rates.
Sources: Cleaveland (2026)
Choice and Response Chunking
Response chunking, the organization of individual responses into larger behavioral units, is directly relevant to how preference is expressed under concurrent schedules. When a COD was applied to one VI 30-s schedule, pigeons emitted clusters of rapid pecks upon entry to that schedule, a pattern absent from the schedule without a COD, and birds were far less likely to switch away during the first four seconds at the COD schedule. Accounting for how such chunks are formed and brought under stimulus control is necessary for reconciling molecular and molar accounts of action selection.
Sources: Cleaveland (2026)
Choice and Discriminative Stimulus Control
Discriminative stimulus control over trained response structure determines whether preference observed during reinforced training transfers to probe conditions. In one experiment, a traditional Findley procedure preserved the stimulus context established during training, and birds showed a clear peck-based preference for the VI 30-s stimulus that had been paired with a COD, whereas a procedure that altered that context produced the opposite result. What appears as preference in a probe test reflects the degree to which the probe reinstates the discriminative conditions under which the response chunks were originally acquired.
Sources: Cleaveland (2026)
