Horses do not differentiate between harsh and soft voice cues
12 Apr 2015
A joint study by researchers in the USA and Europe looked at whether horses respond to harsh versus soft voice cues when completing a frightening task. The researchers hypothesized that horses with the soft voice reinforcement would have a lower heart rate, display calmer behaviour and take less tries to complete a novel task.
The experiment was carried out using 107 horses assigned to either a “harsh voice” or “soft voice” reinforcement treatment, the groups were balanced for age and gender. A handler were given the task of getting a horse to cross a scary object, a tarpaulin. The handler did this by leading a horse in a halter towards the object on the ground. When the horses took a step forward the handler released the pressure on the halter and followed up with a harsh voice or soft voice reinforcement. During the harsh voice treatment the handler told the horse to “quit it” in a loud, sharp manner, during the soft voice treatment the handler told the horses “good horse”in a soft, soothing manner. The two treatments were acoustically verified to be significantly different from each other, similar between different handlers participating in the study and within a horse’s range or hearing.
Contrary to their hypotheses, the researchers did not find any significant differences between the two treatments. Horses with the harsh voice treatment did not show any significant variation in handling behaviour, time taken to complete the task, average heart rate or maximum heart rate.
One explanation of the results favoured by the researchers is that in previous studies, animals, when presented with two simultaneous cues related to a task, only took notice of one. The horses used in this study were extremely familiar with pressure-release training and therefore responded to the release of pressure from the halter and did not show any physiological response to the voice cue, suggesting it either meant nothing to them or they blocked it out.
If this is the case, it does not mean that horses cannot learn to distinguish tones of voice but it may mean that they require classical conditioning to do so. This means that the horses would have to be trained enough times with pressure-release and a consistent voice cue that over time they learn to associate the voice cue with the release of pressure and can go off the voice cue alone without the application of pressure to start with. In a situation such as this a consistent tone of voice cue could still be important.
Implications this that this study could have for real life: if you horse is doing something naughty and you are yelling at him but at the same time releasing pressure, there is a good chance he is going to learn from the pressure release rather than the yelling. He then might be inclined to repeat that behaviour again in the future.
Heleski, C., Wickens, C., Minero, M., DallaCosta, E., Wu, C., Czeszak, E., & von Borstel, U. K. (2014). Do soothing vocal cues enhance horses’ ability to learn a frightening task?. Journal of Veterinary Behavior: Clinical Applications and Research.