Effects of retention interval on performance in a numerical reproduction task

Area of Science:

• Animal Cognition
• Behavioral Neuroscience
• Comparative Psychology

Background:

• Numerical reproduction tasks in animals explore memory and cognitive processes.
• Previous research suggests a "choose-short/small" effect, where shorter retention intervals (RI) lead to choosing smaller numbers.
• The impact of extended retention intervals on numerical reproduction remains less understood.

Purpose of the Study:

• To investigate if increasing the retention interval (RI) in a numerical reproduction task with pigeons elicits a "produce-small" effect.
• To examine the relationship between retention interval duration and numerical response patterns in pigeons.

Main Methods:

• Four pigeons were trained on a numerical reproduction task with a 2-second retention interval.
• Pigeons were subsequently tested with varied retention intervals of 0.5 seconds and 8 seconds.
• Response numbers and accuracy were analyzed across different retention intervals and sample stimuli.

Main Results:

• A number-dependent, "produce-large" effect was observed: response numbers increased with longer retention intervals.
• The retention interval significantly impacted responding, particularly on 2-flash trials with an 8-second RI.
• Discrimination accuracy between trial types decreased as the retention interval increased.

Conclusions:

• The findings contradict the "choose-short/small" effect and support a "produce-large" effect in this context.
• The "produce-large" effect may result from a disruption in stimulus control over responding due to extended retention intervals.
• This study highlights the complex influence of memory delays on numerical cognition in non-human animals.