Skip to main content
Log in

How chimpanzees (Pan troglodytes) perform in a modified emotional Stroop task

  • Original Paper
  • Published:
Animal Cognition Aims and scope Submit manuscript


The emotional Stroop task is an experimental paradigm developed to study the relationship between emotion and cognition. Human participants required to identify the color of words typically respond more slowly to negative than to neutral words (emotional Stroop effect). Here we investigated whether chimpanzees (Pan troglodytes) would show a comparable effect. Using a touch screen, eight chimpanzees were trained to choose between two simultaneously presented stimuli based on color (two identical images with differently colored frames). In Experiment 1, the images within the color frames were shapes that were either of the same color as the surrounding frame or of the alternative color. Subjects made fewer errors and responded faster when shapes were of the same color as the frame surrounding them than when they were not, evidencing that embedded images affected target selection. Experiment 2, a modified version of the emotional Stroop task, presented subjects with four different categories of novel images: three categories of pictures of humans (veterinarian, caretaker, and stranger), and control stimuli showing a white square. Because visits by the veterinarian that include anaesthetization can be stressful for subjects, we expected impaired performance in trials presenting images of the veterinarian. For the first session, we found correct responses to be indeed slower in trials of this category. This effect was more pronounced for subjects whose last anaesthetization experience was more recent, indicating that emotional valence caused the slowdown. We propose our modified emotional Stroop task as a simple method to explore which emotional stimuli affect cognitive performance in nonhuman primates.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others


  1. It should be noted that for the color discrimination training A, the onsets of the trial initiation display, the 500-ms waiting display, the stimuli, and the feedback interval were each accompanied by additional program-execution-related average delays of approximately 1–16 ms. For the color discrimination training B, the onsets of the trial initiation display and of the 500-ms waiting display, the first onset of the stimuli, and the feedback interval onset were each accompanied by additional program-execution-related average delays of approximately 7–16 ms.

  2. Following the suggestion of an anonymous reviewer, we also explored latencies in incorrect trials. Across sessions and categories, while response latencies in our sample of seven subjects tended to be slower in incorrect trials (mean of latency medians: M = 862.00 ms) than in correct trials (M = 768.07 ms), this effect was not statistically significant, t(6) = 1.55, p = .173. We further investigated specifically for trials presenting the veterinarian, whether response latencies from subjects with anaesthetization experience were slower in incorrect than in correct trials: Considering data from all three sessions, we did not find a significant difference between latencies in correct vs. incorrect vet trials, t(5) = 1.43, p = .211. Considering Session 1 alone, in spite of a sizable mean difference in response latency between incorrect vet trials (M = 1320.08 ms) and correct vet trials (M = 908.75 ms), this effect was not statistically significant, t(5) = 1.98, p = .105. We would like to add a note of caution with regard to these results, however. For several subjects, the rate of incorrect responses was very low. In particular, when considering vet trials alone, this means that some of the latency scores that had to be used in these analyses were based on as little as two to four data points (three subjects in Session 1). Such small numbers imply low measurement reliability, even when medians are used as measures of central tendency. For the same reason, latency comparisons between different categories that were restricted to incorrect trials were not carried out because several subjects made no errors in one or more of the non-veterinarian categories in one or more of the three sessions.


Download references


The authors wish to thank the staff at Leipzig Zoo, particularly the zoo veterinarian and the chimpanzee caretakers, for their various contributions to stimulus preparation and data collection. We thank Thurston Cleveland Hicks and Fabrizio Maffessoni for their contributions to stimulus preparation. We thank Alexander Weiss for providing a German version of the Hominoid Personality Questionnaire. We thank Daniel Geissler, Stefan Leideritz, Johannes Grossmann, and Sarah Peoples for providing personality ratings of the chimpanzees.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Matthias Allritz.

Ethics declarations

Animal husbandry and research comply with the “EAZA Minimum Standards for the Accommodation and Care of Animals in Zoos and Aquaria,” the “WAZA Ethical Guidelines for the Conduct of Research on Animals by Zoos and Aquariums,” and the “Guidelines for the Treatment of Animals in Behavioral Research and Teaching” of the Association for the Study of Animal Behavior (ASAB).

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Allritz, M., Call, J. & Borkenau, P. How chimpanzees (Pan troglodytes) perform in a modified emotional Stroop task. Anim Cogn 19, 435–449 (2016).

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: