Toward a Choice-Based Judgment Bias Task for Horses.

Area of Science:

• Animal behavior and welfare science
• Equine cognition and behavior research
• Operant conditioning and animal learning

Background:

• Judgment bias tasks are valuable for assessing animal emotional valence and welfare.
• Establishing valid judgment bias tasks is crucial for equine welfare assessment.
• Auditory cue discrimination is key to understanding animal emotional states.

Purpose of the Study:

• To evaluate two auditory judgment bias task versions (go/no-go and active choice) for horses.
• To assess the acquisition learning and discrimination of ambiguous auditory cues in horses.
• To determine the suitability of these tasks for measuring equine emotional valence.

Main Methods:

• Ten horses (5 mares, 5 stallions) were randomly assigned to either the go/no-go or active choice task.
• Horses were trained on operant responses to low- and high-frequency tones.
• Ambiguous tones were introduced during a testing phase to assess cue discrimination.

Main Results:

• All horses in the go/no-go group learned the task, while only one in the active choice group succeeded.
• The go/no-go task demonstrated easier acquisition learning compared to the active choice task.
• Go/no-go task participants failed to differentiate between ambiguous auditory cues during testing.

Conclusions:

• The go/no-go auditory judgment bias task is more feasible for training horses than the active choice version.
• Current task designs may not be sufficiently valid for assessing equine emotional valence due to poor ambiguous cue discrimination.
• Further refinement of judgment bias tasks is needed to reliably measure emotional states in horses.