Fagen et al. (Elephant Learning)

This study is applied operant conditioning — so master the mechanism and the study explains itself.

Operant conditioning (Skinner) = behaviour is learned through its consequences:

• Positive reinforcement = adding something pleasant after a behaviour → behaviour increases (e.g. giving an elephant food after it lifts its trunk).
• Negative reinforcement = removing something unpleasant → behaviour increases (NOT used as the main method here — the point of the study is to avoid unpleasant/harsh methods).
• Punishment = a consequence that decreases behaviour (traditional elephant training often used punishment — this study offers an alternative).

Primary vs secondary reinforcers:

• Primary reinforcer = satisfies a basic need directly (e.g. food — banana, sugar cane).
• Secondary reinforcer = a neutral signal that becomes rewarding through association with a primary one (e.g. a clicker, whistle or the word 'good'). This is secondary positive reinforcement (SPR) — the trainer can mark the exact moment of correct behaviour, then give food.

Shaping = reinforcing successive approximations — rewarding behaviours that get closer and closer to the target (e.g. first any trunk movement, then trunk to the bucket, then trunk in the bucket).

Behavioural chaining = linking several trained behaviours into a sequence to perform a complex action — here, the full trunk wash is a chain of separate trained steps.

Psychology being investigated: operant conditioning — reinforcement, shaping and chaining.

Background. Working elephants in Nepal can carry tuberculosis (TB), which can spread to humans. Testing for TB requires a trunk wash — the elephant puts its trunk in fluid, holds it, then blows it out so the sample can be collected. Traditional elephant training often used harsh, punishment-based methods. Fagen et al. wanted a humane alternative.

Aim. To investigate whether secondary positive reinforcement (SPR) could be used to train captive working elephants to voluntarily perform a trunk wash, demonstrating that operant conditioning offers a practical, welfare-friendly alternative to traditional training.

Sample: 5 captive working elephants in Nepal, trained with the help of their handlers (mahouts). (Human participants were not the focus; the elephants are the subjects.)

The training method — secondary positive reinforcement (SPR):

• A secondary reinforcer — an acoustic signal such as the word 'good' (or a whistle) — was paired with a primary reinforcer (food, e.g. chopped banana or sugar cane). The signal marks the correct moment; the food rewards it.
• Behaviours were shaped by reinforcing successive approximations, then chained together to build the full trunk wash.

The trunk-wash behaviours trained (the chain) included things like: bringing the trunk here, trunk up, placing the trunk in the bucket, holding steady, and blowing out the fluid — so a sample could be collected for TB testing.

Measuring success: a behavioural checklist recorded whether each elephant could perform each trained behaviour to criterion, and the number of training sessions/time needed — allowing the researchers to judge how successful SPR was for each elephant.

Results:

• Secondary positive reinforcement worked — it successfully trained elephants to perform trunk-wash behaviours, and some elephants learned the full trunk wash voluntarily.
• There were clear individual differences: elephants varied in how quickly they learned and how much of the chain they mastered. Some learned individual behaviours but did not complete the whole sequence to criterion.
• Differences may relate to factors such as the age of the elephant and the handler/relationship — showing learning is affected by individual and situational factors.

Conclusion:

• SPR is a viable, humane training method for working elephants, enabling important health procedures (TB testing) without harsh, punishment-based methods.
• It demonstrates that operant conditioning (shaping + chaining) works in applied, real-world settings and can improve animal welfare.
• Because not all elephants reached full criterion, the method may need adapting to individual animals.

Strengths

• High ecological validity — real working elephants trained in their natural working environment in Nepal, so the behaviour is genuine, not lab-contrived.
• Strong real-world application — provides a humane training method that improves welfare and enables vital TB testing; directly useful.
• Objective measurement via a behavioural checklist → reliable recording of success.

Weaknesses

• Very small sample — only 5 elephants, so findings may not generalise to all elephants (or other species/contexts).
• Field study → less control — many uncontrolled variables (handler differences, individual histories), so cause-and-effect is harder to establish than in a lab.
• Individual differences make standardisation difficult — each elephant/handler pairing differed.

Issues & debates

• Use of animals in research (KC2): here the animals' welfare is the goal, so the ethics are largely positive (the method reduces harm) — but generalisability from 5 elephants is still limited.
• Nature vs nurture: strongly nurture (the behaviour is learned through reinforcement).
• Application to everyday life: humane animal training, zoos, veterinary care.
• Individual differences: elephants varied in learning.