Toward a Choice-Based Judgment Bias Task for Horses.

This research article investigates how judgment bias tasks – tasks that assess emotional responses – could be used for horses to measure animal welfare. The specific aim is to determine the effectiveness of two versions of an auditory judgment bias task, a go/no-go task and an active choice task. Results show that while the go/no-go method was more easily learned by the horses, the efficacy of this method to differentiate between ambiguous auditory cues is questionable.

Understanding the Concept of Judgment Bias

• Judgment bias tasks are typically used in measuring animal welfare by assessing the emotional responses. These tasks are based on the notion that animals, like humans, exhibit positive or negative biases which can be influenced by their emotional states.

Design and Execution of the Experiment

• The researchers used two types of tasks, a go/no-go task and an active choice task. In the go/no-go task, horses would perform an action in response to a specific sound, but remain still for a different sound. In the active choice task, horses were trained to make different responses to low-frequency and high-frequency tones respectively.
• Five mares and five stallions were randomly allocated to one of these two tasks. They were then trained for ten trials daily to learn the required responses for the task.
• Once the horses had learned their assigned task, they were tested over four days with three trials involving ambiguous tones intermingled with the high-tone and low-tone trials.

Findings of the Study

• All the horses assigned to the go/no-go task successfully learned the task. In contrast, only one horse who was assigned the active choice task managed to learn the task. From this result, the experiment concludes that the go/no-go task is easier for horses to learn compared to the active choice task.
• Despite this, the horses taught with the go/no-go task were unable to distinguish between the ambiguous cues during the testing phase. The active-choice horse, however, was able to successfully differentiate these cues, calling into question the validity of using the go/no-go task for assessing emotional valence.

Implications of the Research

• The conclusions derived from this study could have significant implications for how to train and assess horses for welfare purposes. It helps identify the most effective approaches to teach horses to perform specific tasks or responses.
• It also highlights the need to further study and refine our understanding of judgment bias tasks, as not all formulations of these tasks may accurately capture the well-being or emotional states of animals.