Fagen et al LA Flashcards
Psych being investigated-General
➢captive elephants play an important role in Nepalese culture (part of many religious ceremonies and customs + work w timber trade, transportation, park management, tourism.
➢Unfortunately, elephants are vulnerable to TB-potentially fatal disease which can be transmitted between wild elephants + captive + humans…
➢It affects more than 20% of Nepali elephants, meaning finding solutions is crucial to both human and animal welfare.
➢Study investigates a way of trainings the animals to actively participate in a veterinary procedure called a trunk wash-allows them to be tested for disease.
➢Elephants had to learn+perform series of behaviors in correct order.
➢If they weren’t accurately performed TB test result may not be accurate and an elephant could be infected without knowing.
➢This means that elephants must achieve a high standard in every behavior that they’re taught.
Psych being investigated-Operant conditioning + positive reinforcement
➢researchers trained elephants using operant conditioning
➢In this type of learning, the consequence of a behavior determines the probability of that behavior being performed again.
➢Fagen et al trained the elephants by rewarding them for performing specific trunk movements in the correct order. This is called positive reinforcement.
Psych being investigated-Primary+secondary reinforcement
➢Each time elephant moved its trunk into correct position,it was rewarded w a banana.
➢Chopped banana is a primary reinforcer bc it meets a basic need (hunger) + elephant was more likely to move its trunk in the same way again if the movement was rewarded w a banana-a food they really enjoy.
➢Operant conditioning is most successful when the time between the behavior (trunk movement) and the consequence is as brief as possible.
➢For this reason, animal trainers often use secondary reinforcers, such as sounds, which the animal is taught to associate w the primary reinforcer.
➢Here, they used a short blast of a whistle-allowed to reward elephants exactly as desired behavior was performed , creating a strong (sensory) association between behavior + consequence.
➢The whistle is a secondary reinforcer because in itself it does not meet any of the elephants needs, but if the researchers train the elephant to associate the whistle with banana, it can become a powerful reinforcer in itself.
Psych being investigated-shaping + behavioral chaining
🔸In order to perform trunk wash, elephants had to learn separate behaviors.
🔸1)allow trainers to inject saline solution into trunks, 2) Next, had to hold trunks upright so fluid could travel to trunk base. 3) Spray fluid into bucket for testing.
🔸To begin w, animal is rewarded for moving trunk into any position similar to desired position. Gradually, elephant is only rewarded for movements/positions that are increasingly similar to desired behavior.
🔸This is called shaping. Once its mastered first behaviors in trunk-wash sequence,they move on learning other one.
🔸Over time, trainer will begin to reward them only for chaining desired behaviors together (i.e first followed by second…).Once they’ve done this, only rewarded after doing 3 in correct order (i.e perform trunk wash), w each behavior performed to a satisfactory standard.
Background
🔸In 2011, the Nepalese government introduced annual TB of captive elephants to reduce transmission-causing researchers to develop new training methods using positive reinforcement.
🔸Hoped that elephants would participate voluntarily,reducing stress, improving captive animal welfare+protecting trainers from injury due to anxious and un-co-operative elephants.
🔸Similar techs had alr been used w non human primates,antelopes,giant pandas-but first time such techs trialled w elephants.
🔸Elephants in Nepal are generally trained and taed by handlers called mahouts-also use principles of operant conditioning. Traditional training methods use aversive stimuli, e.g being prodded w a sharp bamboo stick called kocha.
🔸If a mahout wants an elephant to turn right but it turns left it might poke w stick as punishment. Sticks also used for negative reinforcement-mahout prods elephants until they turn in correct direction.
🔸Due to concerns over animal welfare+keeper safety, new methods like those described are slowly being introduced, focusing firmly on use of positive reinforcement.
Fagen et al. in brief!
🔸Study explores effectiveness of secondary positive reinforcement for training working elephants to voluntarily participate in trunk washing-a veterinary procedure used to test for TB.
🔸4/5 elephants successfully mastered the necessary behavior within 35 training sessions (av session lengths=12mins).
🔸Training was seen to be effective in reliably producing the desired behavior of trunk washing for (juvenile) infant elephants.
Aim
🔸This observational study aimed to investigate whether secondary positive reinforcement could be used to train the elephants to voluntarily complete a trunk wash, a veterinary behavior that allows them to be tested for TB.
Research Methodology A
🔸Fagen et al conducted a controlled observation of the elephant training sessions. The observations took place while the elephants were chained in their stalls at the stable where they lived and not in the wild, for example.
🔸It was also structured in that the observers used a behavioral checklist to measure how successful elephants were in completing various behaviors involved n trunk wash.
Research Methodology B
Description of Trained Behaviours:
🔸Trunk here
-The distal end of the trunk is placed gently on top of the outstretched palm of the trainer, with the ventral aspect of the trunk in contact with the trainer’s palm.
🔸Trunk up
-The distal end of the trunk is held upward either in a loose curl with the dorsal aspect of the tip of the trunk in close contact with the elephant’s own forehead or is held diagonally up and outward with a completely straight trunk. The exact height or angle of the trunk is not measured.
🔸Bucket
-The distal end of the trunk is gently placed inside a bucket.
🔸Blow
-The elephant gives a strong, sharp exhale through the trunk.
🔸Steady
-The elephant holds the trunk still with the trunk held in the position previously requested (trunk here, trunk down or trunk out). The elephant can move his or her feet, ears, head, tail and body slightly as long as the trunk remains still in the previous position requested.
🔸Syringe
-The elephant holds the trunk still in the trunk-here position to have the distal end of a catheter tip syringe placed inside the nostril of the trunk and up to 60 ml of saline or water instilled into the trunk.
🔸Blow into bucket
-The elephant places the distal end of the trunk in the bucket and gives a strong, sharp exhale through the trunk.
🔸Trunk down
-The trunk is held in a relaxed position with the trunk hanging loose towards the ground.
🔸Trunk out
-The trunk is held stretched straight outward, approximately parallel to the ground.
🔸Targeting
-The elephant moves such that the centre of the forehead makes contact with the end of a targeting stick placed at the height of the forehead.
Research Methodology-Sample
🔵 The researchers studied five female elephants – four juveniles (infants) aged 5–7 years old and one adult.
🔵 The young elephants were all born in captivity.
🔵 The adult elephant was at least 50 years old.
🔵 The elephants were chosen as they were all relatively gentle/tame and none was pregnant, and because their mahouts (handlers) were all happy to take part.
🔵 The elephants had no previous experience with secondary positive reinforcement training and had only ever been trained using traditional methods.
🔵 The mahouts were present in all sessions to maintain the trainers’ safety but they did not interact with the elephants.
Procedure-Training elephants
🔸Elephants were trained in morning (7:30-10.00am) and/or afternoon (4-7pm).
🔸Session times depended on availability of mahouts, but were never more than 2 days apart.
Procedure-What did elephants have to do?
To complete trunk wash, elephants had to:
1.Put trunk into trainers hand(so saline/sterile water can be inserted)
2.Lift trunk and hold (so fluid can flow to base)
3.Lower trunk into bucket
4.Blow into bucket to remove fluid.
5.Hold steady (keep trunk still)-a final behavior marking end of sequence, before elephants can relax.
Initially, the syringe of saline/sterile water was not used; the elephants just had to learn the different behaviours that would be necessary.
Three other steps called targeting, trunk down, and trunk out were also taught but not included in this study, as they were not part of the trunk wash.
Procedure-How were elephants trained?
Capturing – The elephants were encouraged to perform natural behaviours that were already part of their usual routine, using rewards – for example, if an elephant happened to hold her trunk up, she would be given chopped banana (a primary reinforcer); this captured the behaviour and made trunk lifting more common.
Luring – More unusual behaviours were encouraged by careful positioning of treats by the trainer – for example, if the trainer needed the elephant to stretch its trunk out, they would put a treat out of reach to lure them towards it.
Shaping – To begin with, elephants were rewarded for all behaviours that were similar to the final goal – for example, they might be rewarded for blowing fluid near to the bucket. However, gradually the elephants were only rewarded for more accurate behaviours – for example, blowing into the bucket. Their behaviours were shaped to become more accurate over time.
Secondary reinforcers – The trainers taught the elephants to associate the sound of a whistle (a secondary reinforcer) with the arrival of chopped banana (a primary reinforcer). This meant the whistle could be used to reward behaviour immediately, helping the elephant to quickly learn the association between their behaviour and the reward.
– encouraged them to accept the syringe being placed inside the trunk, slowly moving towards injecting a droplet of fluid, gradually from 1 to 15 ml, to the full 60 ml necessary for the test.
Procedure-Verbal cues and behavioral chaining
➡Trainers introduced one syllable verbal cues to prompt elephants once they had successfully learnt all 5 behaviors.
➡These cues were non-words and meant nothing in English or Nepali as the researchers did not want the mahouts to think the elephants could understand human language (common myth in area).
➡ Behavioral chaining was used to encourage elephants to perform 5 behaviors in order-following the verbal cue.
➡Starting w just 2 behaviors, the elephants were only rewarded if they completed the behaviors in the correct order.
➡Gradually, more behaviors were added until elephant could perform all 5 in order.
Procedure-The syringe and sample fluid
🔸Syringe was only introduced when the elephants had learnt the trunk wash behaviors
🔸Syringe is an aversive stimulus as elephants might find it unpleasant.
🔸To help them accept it, trainers gradually brought it closer to trunk, during step1 (trunk in hand), rewarding elephants w banana, until they were happy to have syringe touch hands (great as it wasnt immdeiately forced imo).
🔸This is called desensitization.
🔸Researchers also used counter conditioning to teach elephants to associate syringe with arrival of chopped banana.
🔸Gradually, the syringe went from being an aversive stimulus to a conditioned stimulus.
🔸Once elephants were able to tolerate syringe touching outside of trunks, the trainer gradually encouraged them to accept it being placed in trunk, slowly moving towards injecting a droplet of fluid, gradually building from 1-15ml, to full 60ml necessary for test.
Procedure-Measured variables
There were 3 main measured variables recorded by an assistant who observed each training session:
🔵mins of training from the point at which the elephant was offered her first cue to her response to the last cue
🔵num of offers/cues made by trainer to elephant
🔵success rate for each behavior and each sequence, starting at session 10, the elephants were tested after every 5th session. They were tested on everything they had learned up to that point.
-They were considered to have passed the training if they showed the correct behavior following 8/10 offers/cues (80%). Behaviors were judged as successful if the trainer felt that they would function adequately in a real TB trunk-wash test.
-Elephants were also graded on their ability to perform sequences of behaviours – e.g, trunk in hand followed by trunk up. The success criterion was again 80 per cent and they could score between 80 and 100 per cent on any sequences they had learned.
-In this case, individual behaviours were not retested as the elephant had already shown them as part of the sequence. In this case, the individual behaviours were scored as 90 per cent successful.
-If an elephant failed a sequence, they were tested on shorter sequences and/or individual behaviours.
-The only two behaviours that were regularly tested on their own were steps 3 and 5: trunk down and hold steady.
-Training was concluded when the elephant had a success rate of ≥ 80 per cent on the full five-step trunk wash.
Results A
-The four juvenile elephants all learned the full trunk wash in 25–35 sessions.
-The mean average session duration was 12 minutes, ranging from 10 to 13 minutes, and the overall training time for these elephants was 367 minutes, with a range of 194 minutes (257–451 minutes).
-Elephant 5, the older elephant, was never tested on the full trunk wash as she failed to learn the full sequence in the time available.
-The behaviours that she could not master were blowing into the bucket and holding her trunk steady.
-Elephants 2 and 4 also never fully mastered the trunk steady behaviour, except as part of the full trunk wash.
-Elephant 5 was also never fully desensitised to the syringe. When her training sessions were added, the mean increased to 378 minutes.
-Over the course of the study, the elephants’ success rate for accurate individual behaviours and sequences increased from 39 per cent after 10 sessions to 89.3 per cent after 35 sessions.
Results B
Data on number of sessions and average session times for individual elephants
🔹 Elephant 2
-Number of sessions to pass trunk wash: 25
-Mean average session time (minutes): 10.29
-Individual behaviours failed: Steady
🔹 Elephant 1
-Number of sessions to pass trunk wash: 30
-Mean average session time (minutes): 12.42
-Individual behaviours failed: None
🔹 Elephant 3
-Number of sessions to pass trunk wash: 35
-Mean average session time (minutes): 13.27
-Individual behaviours failed: None
🔹 Elephant 4
-Number of sessions to pass trunk wash: 35
-Mean average session time (minutes): 11.11
-Individual behaviours failed: Steady
Results C
🔵 Individual behaviours varied significantly in terms of the amount of practice necessary to master them.
🔵 For example, the most difficult behaviour was the first one (trunk here/trunk in hand), which took an average of 295 offers,
🔵 compared with lower trunk into bucket (61 offers) and blow into bucket (54 offers), which required considerably less practice (see Figure 3.11).
- Figure: Average offers made for each of the basic behavioural tasks required to pass a performance test, indicating relative difficulty in learning the task.
- Total HFUBBs represents the total offers made for the trunk-wash sequence (here, fluid, up, bucket, and blow).
🔵 Total HFUBBs – approx. 175 offers
🔵 Desensitisation – approx. 165 offers
🔵 Blow in bucket – approx. 55 offers
🔵 Blow – approx. 210 offers
🔵 Bucket – approx. 60 offers
🔵 Trunk here – approx. 295 offers
🔵 Trunk up – approx. 150 offers
Evaluation-Ethical issues-Distress
One strength is that ethical guidelines were followed for both animals and trainers. Despite elephants’ leg chains, they were able to walk away if they did not want to participate. The chains still allowed them to move freely within their stalls. This is important as it helped reduce distress (psychological harm) of the animals asw as risk of physical harm to trainers, observers, and/or mahouts, who could have been injured if an elephant became too unco-operative.
Evaluation-Methodological issues-Reliability- Using a behavioral checklist
-Another strength was use of behavioral checklist which detailed the exact operational descriptions of each of the taught behaviors.
-E.g, the trunk up behavior was described as “the distal end of the trunk is held upright either in a loose curl w the dorsal aspect of the tip of the trunk in close contact with the elephant’s own forehead or is held diagonally up and outward w a completely straight trunk.” These detailed descriptions (very clearly operationalized) helped increase reliability of the observer’s decisions regarding whether the elephant passed or failed each aspect of the performance test.
Evaluation-Methodological issues-Validity-No additional cues from the mahouts
-strength is that mahouts complied w researchers’ requests to not speak English nor signal to elephants in any way.
-important as it meant that changes in the animals behaviors must have resulted from secondary positive reinforcement training delivered by the trainer and not from additional communication from mahout.
-This helps establish a cause effect relationship and to increase validity of findings that secondary positive reinforcement training is helpful in training elephants to voluntarily participate in trunk washing.
Evaluation-Methodological issues-Validity-Problem with behavioral checklist
A weakness was Fagen et al.’s operational definition of ‘steady’.
To pass the steady performance test, the elephants had to stay still in various different trunk positions. They had to pass three trunk-here and trunk-down tests and four trunk-up tests (a very important part of the trunk wash).
The elephants were better at trunk up and trunk here but not so good at trunk down, only reaching about 70 per cent accuracy overall.
As the success criterion was 80 per cent, the elephants regularly failed the steady task. This is important as the very prescriptive operational definition meant the data did not reflect the elephants’ true ability on two of the trunk positions.
🔹 Another weakness was that the total training time included training on a few behaviours that were not part of the trunk wash.
For example, four elephants were trained on a behaviour called trunk out, but the trainers stopped shaping this behaviour when they realised it was not necessary.
This means that the total training time does not accurately reflect the time taken to train the elephants to perform the trunk wash.
The time would be reduced if the unnecessary trunk-out training sessions were subtracted from the total.
This reduces the validity of the findings.
Evaluation-Methodological issues-Objectivity+Subjectivity-Rating success and failure
A weakness was that the measurement of the elephant’s performance was subjective.
Despite the detailed behavioural checklist, the trainer had to decide whether they felt the elephant would be successful if they were performing a real trunk wash, not just whether it matched the description on the checklist or not.
This is a weakness because the trainers may have been biased towards viewing behaviours as successful when another observer might have interpreted the same behaviour as unsuccessful – for example, the angle of the trunk may not have been quite right or some droplets of fluid may have been spilled.
