How task format affects cognitive performance: a memory test with two species of New World monkeys

Michele N. Schubiger, Alexandra Kissling, Judith M. Burkart

Research output: Contribution to journalArticle

6 Citations (Scopus)
50 Downloads (Pure)

Abstract

In cognitive tests, animals are often given a choice between two options and obtain a reward if they choose correctly. We investigated whether task format affects subjects' performance in a physical cognition test. In experiment 1, a two-choice memory test, 15 marmosets, Callithrix jacchus, had to remember the location of a food reward over time delays of increasing duration. We predicted that their performance would decline with increasing delay, but this was not found. One possible explanation was that the subjects were not sufficiently motivated to choose correctly when presented with only two options because in each trial they had a 50% chance of being rewarded. In experiment 2, we explored this possibility by testing eight naïve marmosets and seven squirrel monkeys, Saimiri sciureus, with both the traditional two-choice and a new nine-choice version of the memory test that increased the cost of a wrong choice. We found that task format affected the monkeys' performance. When choosing between nine options, both species performed better and their performance declined as delays became longer. Our results suggest that the two-choice format compromises the assessment of physical cognition, at least in memory tests with these New World monkeys, whereas providing more options, which decreases the probability of obtaining a reward when making a random guess, improves both performance and measurement validity of memory. Our findings suggest that two-choice tasks should be used with caution in comparisons within and across species because they are prone to motivational biases.
Original languageEnglish
Pages (from-to)33-39
Number of pages7
JournalAnimal Behaviour
Volume121
DOIs
Publication statusPublished - 13 Sep 2016

Fingerprint

Cebidae
cognition
Callitrichidae
testing
Saimiri sciureus
Callithrix jacchus
Saimiri
animal tests
experiment
monkeys
food
test
world
cost
duration

Cite this

Schubiger, Michele N. ; Kissling, Alexandra ; Burkart, Judith M. / How task format affects cognitive performance : a memory test with two species of New World monkeys. In: Animal Behaviour. 2016 ; Vol. 121. pp. 33-39.
@article{ba4d910d63034f2d8d527fedb6f489a4,
title = "How task format affects cognitive performance: a memory test with two species of New World monkeys",
abstract = "In cognitive tests, animals are often given a choice between two options and obtain a reward if they choose correctly. We investigated whether task format affects subjects' performance in a physical cognition test. In experiment 1, a two-choice memory test, 15 marmosets, Callithrix jacchus, had to remember the location of a food reward over time delays of increasing duration. We predicted that their performance would decline with increasing delay, but this was not found. One possible explanation was that the subjects were not sufficiently motivated to choose correctly when presented with only two options because in each trial they had a 50{\%} chance of being rewarded. In experiment 2, we explored this possibility by testing eight na{\"i}ve marmosets and seven squirrel monkeys, Saimiri sciureus, with both the traditional two-choice and a new nine-choice version of the memory test that increased the cost of a wrong choice. We found that task format affected the monkeys' performance. When choosing between nine options, both species performed better and their performance declined as delays became longer. Our results suggest that the two-choice format compromises the assessment of physical cognition, at least in memory tests with these New World monkeys, whereas providing more options, which decreases the probability of obtaining a reward when making a random guess, improves both performance and measurement validity of memory. Our findings suggest that two-choice tasks should be used with caution in comparisons within and across species because they are prone to motivational biases.",
author = "Schubiger, {Michele N.} and Alexandra Kissling and Burkart, {Judith M.}",
year = "2016",
month = "9",
day = "13",
doi = "10.1016/j.anbehav.2016.08.005",
language = "English",
volume = "121",
pages = "33--39",
journal = "Animal Behaviour",
issn = "0003-3472",
publisher = "ACADEMIC PRESS LTD- ELSEVIER SCIENCE LTD",

}

How task format affects cognitive performance : a memory test with two species of New World monkeys. / Schubiger, Michele N.; Kissling, Alexandra; Burkart, Judith M.

In: Animal Behaviour, Vol. 121, 13.09.2016, p. 33-39.

Research output: Contribution to journalArticle

TY - JOUR

T1 - How task format affects cognitive performance

T2 - a memory test with two species of New World monkeys

AU - Schubiger, Michele N.

AU - Kissling, Alexandra

AU - Burkart, Judith M.

PY - 2016/9/13

Y1 - 2016/9/13

N2 - In cognitive tests, animals are often given a choice between two options and obtain a reward if they choose correctly. We investigated whether task format affects subjects' performance in a physical cognition test. In experiment 1, a two-choice memory test, 15 marmosets, Callithrix jacchus, had to remember the location of a food reward over time delays of increasing duration. We predicted that their performance would decline with increasing delay, but this was not found. One possible explanation was that the subjects were not sufficiently motivated to choose correctly when presented with only two options because in each trial they had a 50% chance of being rewarded. In experiment 2, we explored this possibility by testing eight naïve marmosets and seven squirrel monkeys, Saimiri sciureus, with both the traditional two-choice and a new nine-choice version of the memory test that increased the cost of a wrong choice. We found that task format affected the monkeys' performance. When choosing between nine options, both species performed better and their performance declined as delays became longer. Our results suggest that the two-choice format compromises the assessment of physical cognition, at least in memory tests with these New World monkeys, whereas providing more options, which decreases the probability of obtaining a reward when making a random guess, improves both performance and measurement validity of memory. Our findings suggest that two-choice tasks should be used with caution in comparisons within and across species because they are prone to motivational biases.

AB - In cognitive tests, animals are often given a choice between two options and obtain a reward if they choose correctly. We investigated whether task format affects subjects' performance in a physical cognition test. In experiment 1, a two-choice memory test, 15 marmosets, Callithrix jacchus, had to remember the location of a food reward over time delays of increasing duration. We predicted that their performance would decline with increasing delay, but this was not found. One possible explanation was that the subjects were not sufficiently motivated to choose correctly when presented with only two options because in each trial they had a 50% chance of being rewarded. In experiment 2, we explored this possibility by testing eight naïve marmosets and seven squirrel monkeys, Saimiri sciureus, with both the traditional two-choice and a new nine-choice version of the memory test that increased the cost of a wrong choice. We found that task format affected the monkeys' performance. When choosing between nine options, both species performed better and their performance declined as delays became longer. Our results suggest that the two-choice format compromises the assessment of physical cognition, at least in memory tests with these New World monkeys, whereas providing more options, which decreases the probability of obtaining a reward when making a random guess, improves both performance and measurement validity of memory. Our findings suggest that two-choice tasks should be used with caution in comparisons within and across species because they are prone to motivational biases.

U2 - 10.1016/j.anbehav.2016.08.005

DO - 10.1016/j.anbehav.2016.08.005

M3 - Article

VL - 121

SP - 33

EP - 39

JO - Animal Behaviour

JF - Animal Behaviour

SN - 0003-3472

ER -