Is Food or Affection Better as a Reward in Horse Training?
Twenty horses took part in the study. They were Konik
horses, a primitive breed in Poland (pictured). Twelve of the horses were raised in
typical domestic conditions, while the remaining eight were raised in a forest
reserve in semi-natural conditions. At the time of the study, all of the horses
were between 1 and 2 years old, and live in stables.
The horses were taught a ‘stay’ command starting at 5s and
increasing gradually up to 60s if they progressed that far. Training took place
for 5 minutes a day over a six-day period, and was conducted in the middle of
the stables. Loudspeakers playing white noise were used to ensure that the
other horses in the stables could not hear the training take place. After each
training session, the horses were taken to a paddock and given time to run
around.
Ten of the horses were in a ‘food reward’ group. They were
given a piece of carrot each time they correctly responded to the trainer’s
instruction. The other ten horses were in a ‘grooming reward’ group. If they
responded correctly, the trainer scratched them three times on the withers.
The horses in the food reward group learned the stay command
significantly faster than the horses in the grooming reward group. While 90% of
the horses trained with food were able to stay for a minute at the end of
training, only 40% of the horses trained with grooming could do this. In fact
the mean duration of a stay for the grooming-reward horses was 32 seconds.
What happened is that while the grooming-reward horses made
progress in the first two days, after this the rate of progress stalled. On the
other hand, horses rewarded with food made the most progress in the first three
days, but still continued to improve after that.
The scientists also conducted a test to see if the type of
reward affected how much time a horse would spend in close proximity to the
trainer. This is called a “motionless human test.” Before the training, there
was no difference between the two groups of horses. After training, those in
the food reward group approached the trainer more quickly, and spent more time
near her, than those in the grooming group.
This study shows that for horses, just as with dogs,
training is better facilitated by food; physical touch does not produce the
same results.
The authors point out that food also plays an important role
in human relationships, saying “Don’t we also say that little gifts keep
friendship warm? Is there a better little gift than a box of sweets or
chocolates to make a lover’s heart melt or fill a grandmother with joy?”
What kind of foodie gifts do you like to receive?
Reference
Sankey, C, Henry, S, Górecka-Bruzda, A, Richard-Yris, M-A, & Hausberger, M (2013). The way to a man's heart is through his stomach: What about horses? PLoS ONE, 5 (11) DOI: 10.1371/journal.pone.0015446
So glad people are catching on to positive reward training with animals.
ReplyDeleteThis pretty much puts Monty Roberts back in his place, thankfully. I'm referring to a comment he's made so many times; "Using treats to reward a horse in training is very poor horsemanship". No, Mr. Roberts, horses don't understand your tough love theory, and as 45 years as a compassionate trainer, I've only VERY RARELY had a horse bite me...and because I understood her behavior in just learning to accept treats, this particular action by her was a pure accident, and the bite wasn't painful at all.....additionally, she never repeated that single accidental "bite". Heather Mills / HawkFeather Equine Performance Horses
ReplyDelete