Attention Bias: A novel method to assess psychological well-being in group-housed non-human primates.


This research aims to develop and validate a new measure of the psychological state of captive group-housed non-human primates as a means for improving their welfare.


A better understanding of emotions and feelings in animals is critical to animal welfare science. Psychologically healthy animals will be happier, healthier, less stressed, more likely to learn protocols, and provide robust and reproducible data. However, measuring psychological well-being in animals is intrinsically difficult and existing indicators of well-being focus on indirect physiological or behavioural measures that have a number of limitations.

This project will adapt, refine and validate a state-of-the-art welfare assessment tool, based on 'attention bias' (AB), that is the tendency to pay attention to emotionally dominant stimuli and ignore other stimuli, associated with vulnerability to clinical levels of anxiety, and impaired psychological wellbeing.

Research details and methods

Measuring attention bias, linked to underlying emotional states in both humans and captive primate, will be used as the basis of the test.

The project builds upon previous work where individually-housed captive rhesus macaques are presented images of neutral and aggressive macaque faces and their gaze towards either of these images is filmed, or otherwise observed, to measure their AB; their tendency to focus on stimuli (aggressive faces) that reinforces their negative mood.

This will be developed for use with group-housed macaques to provide an easily applied tool that can be implemented by care staff to assess psychological wellbeing in animals living in social groups. The approach will then be validated as a measure of psychological state by triangulating AB scores with behavioural and physiological state indicators and genetic trait indicators of emotion and wellbeing.

Aim: The aim of this pilot study is to contribute to the NC3Rs objective of refinement by developing and validating a state-of-the-art technological tool for assessing psychological wellbeing in captive non-human primates. This novel method focuses on quantifying attention bias - a bias in visual attention that has been linked to underlying emotional state in both humans and (recently by our research group) captive rhesus macaques.


1. Adapt methods we recently developed to assess attention bias in captive and free-ranging male rhesus macaques, for use with group-housed female rhesus macaques.

2. Validate attention bias as a measure of psychological state, by triangulating attention bias scores with behavioural and physiological state indicators, and genetic trait indicators of emotion and wellbeing.

3. Develop a 'quick and easy' attention bias assessment tool that can be applied widely and inexpensively by care staff, to reliably determine individual attention bias scores.

4. Assess the extent to which baseline attention bias scores predict the nature and degree of change in behavioural, physiological and attentional responses of animals to (preexisting) stressful husbandry procedures.

Methodology: Primarily, we will film eye gaze to social stimuli for blind coding from video (Objective 1), and trial development of a flash card assessment tool using remote off-the shelf computer eye-tracking technology (Objective 3). To address Objectives 2 and 4, we will conduct focal animal behavioural observations, enzyme immunoassay (EIA) of salivary hormones and genetic analyses of 5-HTTLPR serial repeat polymorphisms.

Scientific opportunities: The assessment tool will provide staff at research facilities with an open access resource to maintain high standards of welfare. Breeding and supplying psychologically healthy animals to research facilities provides a grass-roots approach to improving quality of science conducted using these animals.

Howarth E, Kemp C, Thatcher H, et al. (2020) Developing and validating attention bias tools for assessing trait and state affect in animals: a worked example with Macaca mulattaApplied Animal Behaviour Science234, [105198]. doi: 10.1016/j.applanim.2020.105198

Kemp C et al. (2017). A protocol for training group-housed rhesus macaques (Macaca mulatta) to cooperate with husbandry and research procedures using positive reinforcement. Applied Animal Behaviour Science 197:90-100. doi: 10.1016/j.applanim.2017.08.006

Bethell EJ et al. (2016). Emotion evaluation and response slowing in a non-human primate: New directions for cognitive bias measures of animal emotion? Behavioral Sciences 6(1):2. doi: 10.3390/bs601000

Bethell EJ (2015). A How-To Guide for Designing Judgment Bias Studies to Assess Captive Animal Welfare. Journal of Applied Animal Welfare Science 18 Suppl 1:S18-42. doi: 10.1080/10888705.2015.1075833

Back to top
Pilot study grant



Principal investigator

Dr Emily Bethell


Liverpool John Moores University


Dr Stuart Semple
Dr Amanda Holmes
Professor Ann MacLarnon

Grant reference number


Award date

Jan 2014 - Apr 2015

Grant amount