Behaviour modification in custody applies behaviourist learning principles to change how offenders behave within prisons and other secure institutions. The approach focuses on observable behaviour and uses reinforcement to increase desirable actions, such as following rules, completing work and avoiding aggression.
The best-known application is the token economy. Offenders earn tokens for agreed target behaviours and exchange those tokens for valued rewards or privileges. However, although token economies can improve behaviour inside custody, this does not necessarily mean that offenders have been rehabilitated.
Behaviourist Principles Behind Behaviour Modification
Behaviour modification is based on the behaviourist approach and, in particular, operant conditioning.
Behaviourists such as B. F. Skinner argued that behaviour is shaped by its consequences. If behaviour is followed by a rewarding consequence, it becomes more likely to happen again. In contrast, behaviour followed by punishment or the loss of something valued becomes less likely.
In custody, staff can therefore attempt to increase desirable behaviours by reinforcing them.
Examples of desirable target behaviours include:
- Attending education sessions.
- Completing assigned work.
- Following prison rules.
- Maintaining personal hygiene.
- Keeping accommodation clean.
- Cooperating with prison officers.
- Interacting appropriately with other prisoners.
- Avoiding aggressive behaviour.
Behaviour modification focuses on observable behaviour rather than attempting to directly change thoughts or emotions.
This makes the approach different from therapies such as cognitive behavioural interventions, which attempt to change the thinking patterns associated with offending.
Positive Reinforcement
Positive reinforcement involves adding a desirable consequence after a behaviour.
For example, an offender may receive a token after attending an education session and completing their work.
The token acts as a reward.
As a result, attending education becomes more likely to happen again.
The process can be represented as:
Desirable behaviour → Reinforcement → Behaviour becomes more likely
For reinforcement to work effectively, it should normally occur soon after the desired behaviour.
If too much time passes between the behaviour and the reward, the connection between them may become weaker.
Negative Reinforcement
Negative reinforcement is often confused with punishment.
However, negative reinforcement actually increases behaviour.
It works by removing something unpleasant after the desired behaviour occurs.
For example, a restriction might be reduced following a sustained period of appropriate behaviour.
The removal of the unpleasant situation reinforces the behaviour that occurred before it.
Therefore:
Negative reinforcement increases behaviour.
In contrast:
Punishment decreases behaviour.
Token Economies
A token economy is one of the most common forms of behaviour modification used in institutions.
Offenders receive tokens, points or credits when they demonstrate agreed target behaviours.
These tokens can then be exchanged for rewards or privileges.
The token itself usually has little value. Instead, it becomes valuable because offenders learn that it can be exchanged for something desirable.
Therefore, tokens are known as secondary reinforcers.
The actual rewards are known as back-up reinforcers.
Possible back-up reinforcers might include:
- Additional recreational activities.
- Greater choice of activities.
- Access to entertainment.
- Preferred leisure opportunities.
- Other appropriate institutional privileges.
Basic rights should not be treated as rewards.
Food, healthcare, safety and other essential needs should not depend on an offender earning tokens.
How a Token Economy Works
A token economy usually follows several stages.
1. Identify target behaviours
Staff decide exactly which behaviours they want to increase.
These behaviours must be clearly defined and observable.
For example:
“Attend all scheduled education sessions.”
This is clearer than:
“Have a better attitude.”
2. Establish a baseline
Staff may observe the offender’s behaviour before introducing the programme.
This provides a baseline against which later improvement can be measured.
3. Decide how tokens will be earned
Specific behaviours are linked to specific numbers of tokens.
For example:
Completing an education session = one token.
Completing assigned work = one token.
Completing a full day without aggressive behaviour = two tokens.
4. Deliver reinforcement
Tokens should be awarded consistently when the target behaviour occurs.
Immediate reinforcement strengthens the association between the behaviour and its consequence.
5. Exchange tokens for rewards
Offenders can exchange their accumulated tokens for agreed rewards or privileges.
The reward gives the token its value.
6. Monitor progress
Staff record whether target behaviours increase.
The programme can then be adjusted if necessary.
7. Reduce dependence on tokens
Ideally, token rewards should gradually become less important.
Eventually, naturally occurring rewards such as praise, achievement, improved relationships and personal satisfaction should help maintain the behaviour.
This process is important because offenders will not have access to a token economy after release.
Shaping Behaviour
Sometimes an offender may initially struggle to demonstrate the final target behaviour.
In this situation, staff can use shaping.
Shaping involves reinforcing gradual improvements towards the desired behaviour.
For example, imagine an offender regularly becomes aggressive during disagreements.
Initially, they might earn reinforcement for walking away from a confrontation rather than becoming violent.
Later, reinforcement might only be given if they use appropriate communication to resolve the disagreement.
Therefore, progressively higher standards are rewarded.
The Importance of Consistency
Token economies depend heavily on staff applying the programme consistently.
If one prison officer awards a token for a behaviour but another officer does not, the offender receives mixed messages.
Therefore, staff need clear guidance about:
- Which behaviours should be rewarded.
- How many tokens should be given.
- When reinforcement should occur.
- Which rewards tokens can purchase.
- What happens when rules are broken.
Poor staff consistency can reduce the effectiveness of the programme.
Individual Differences in Reinforcement
A reward only works if the individual actually values it.
Different offenders may therefore respond differently to the same reinforcer.
For example, access to a particular recreational activity may strongly motivate one offender but have little value for another.
Effective programmes therefore need to consider individual differences when selecting appropriate rewards.
Hobbs and Holt (1976)
Researcher and year: Tom R. Hobbs and Michael M. Holt, 1976.
Aim
Hobbs and Holt aimed to investigate whether a token economy could increase appropriate behaviour among adolescent offenders living within a correctional institution.
They also wanted to examine whether improvements could be maintained over an extended period.
Procedure or Method
The study involved 125 adolescent males who had been placed in a state correctional institution.
The young offenders lived across four cottage units.
A token economy was introduced into three of the cottages.
However, it was introduced at different times.
The fourth cottage acted as a comparison condition.
This created a multiple-baseline design.
The token programme rewarded behaviours such as:
- Appropriate interactions with peers.
- Following institutional rules.
- Completing assigned tasks.
The researchers collected behavioural data over approximately 14 months.
Findings
Appropriate behaviour increased after the token economy was introduced.
Importantly, behaviour improved in each cottage after the programme began there.
The comparison cottage did not demonstrate the same programme-related pattern of improvement.
Conclusion
Hobbs and Holt concluded that token reinforcement could increase desirable behaviour among adolescent offenders in a correctional institution.
The findings therefore support the use of operant conditioning as a method of changing behaviour in custody.
Why Hobbs and Holt Matters
The study provides useful evidence because the token economy was introduced at different times across different cottages.
Behaviour improved when the intervention was introduced.
Therefore, it becomes more difficult to explain the improvements simply through the passage of time.
However, the study mainly demonstrated improved behaviour inside the institution.
It did not establish that offenders were less likely to reoffend after release.
Ayllon and Milan (1979)
A-Level resources sometimes refer to this research as Ayllon et al. (1979).
However, the commonly cited 1979 correctional behaviour modification work is more accurately associated with Teodoro Ayllon and Michael Milan.
Aim
Ayllon and Milan aimed to investigate and develop ways in which behaviourist principles could be used to manage and rehabilitate offenders within correctional environments.
Procedure or Method
Ayllon and Milan’s work brought together behaviour modification procedures that had been used within correctional settings.
Therefore, it is better viewed as part of a wider programme of correctional behaviour modification research rather than one simple laboratory experiment.
Behavioural programmes identified specific desirable behaviours.
These included behaviours such as:
- Maintaining personal hygiene.
- Keeping cells or living areas appropriately maintained.
- Following institutional rules.
- Participating in education.
- Completing assigned activities.
Offenders received reinforcement when these target behaviours occurred.
Token economies were one method used to deliver reinforcement.
Findings
The research associated with Ayllon and Milan showed that reinforcement could increase specific desirable behaviours while the reinforcement programme was operating.
Improvements were reported in areas such as:
- Rule compliance.
- Hygiene.
- Daily routines.
- Educational behaviour.
However, some behaviour declined again after token systems were removed.
Conclusion
Behaviour modification can successfully change clearly defined behaviour inside correctional institutions.
However, the evidence is stronger for short-term institutional behaviour change than for long-term rehabilitation.
Behaviour Management or Rehabilitation?
One of the most important issues surrounding token economies is the difference between managing behaviour and rehabilitating offenders.
An offender may follow rules because they know that good behaviour will earn tokens.
However, this does not necessarily mean that their underlying attitudes or motivations have changed.
For example, offending behaviour outside prison may be influenced by:
- Criminal thinking patterns.
- Peer pressure.
- Substance misuse.
- Emotional regulation difficulties.
- Relationships.
- Social circumstances.
- Financial pressures.
- Opportunities to offend.
A token economy may not directly address these factors.
Therefore, an offender could become very skilled at following institutional rules while still remaining at risk of reoffending after release.
Reflection question: If an offender behaves well only because tokens are available, has the offender genuinely changed, or has the institution simply changed the consequences surrounding their behaviour?
Evaluation
Strengths
Research Support
One strength of behaviour modification is that research supports its effectiveness.
Hobbs and Holt (1976) found that appropriate behaviour increased after token economies were introduced into three cottage units containing adolescent offenders.
The use of a comparison cottage strengthens this evidence because improvement was linked to the introduction of reinforcement.
Therefore, token economies are supported by empirical evidence rather than being based only on theoretical assumptions.
However, the evidence mainly demonstrates changes in behaviour within custody rather than long-term reductions in offending.
Objective Measurement
Another strength is that target behaviours can be clearly operationalised and measured.
For example, researchers can record:
- Number of education sessions attended.
- Number of tasks completed.
- Number of aggressive incidents.
- Number of rule violations.
Behaviour can therefore be measured before and after a programme begins.
This increases scientific credibility because researchers can evaluate whether behaviour has objectively changed.
Practical Application
Token economies can also have important practical benefits.
Prison officers and other staff can reinforce desirable behaviour during ordinary daily routines.
This means reinforcement does not necessarily require lengthy individual therapy sessions.
If token economies increase cooperation, reduce aggression and encourage participation in education, they may make prisons safer and easier to manage.
Therefore, even if they do not completely rehabilitate offenders, they may still provide valuable institutional benefits.
Relatively Straightforward to Administer
Once target behaviours and rewards have been established, the system can be relatively straightforward for trained staff to administer.
This makes token economies potentially more economical than intensive individual psychological therapies.
Many offenders can participate simultaneously.
Therefore, behaviour modification may offer a relatively efficient way of managing behaviour across a large institution.
However, effective staff training remains important.
Weaknesses
Limited Generalisation
One major weakness is that desirable behaviour may not continue after the token economy ends.
Inside custody, behaviour is controlled within a highly structured environment.
Offenders receive clear and immediate consequences for their actions.
However, once released, these tokens and institutional rewards disappear.
Therefore, behaviour learned in custody may fail to generalise to the outside world.
Some token economy research has found that behaviour begins to return towards previous levels once reinforcement is removed.
This suggests that token economies may create short-term compliance rather than genuine long-term rehabilitation.
Reductionism
Behaviour modification can also be criticised for being reductionist.
It explains behaviour largely in terms of environmental consequences and reinforcement.
However, offending is much more complex.
Criminal behaviour may also involve:
- Cognitions.
- Emotions.
- Biological factors.
- Social relationships.
- Peer groups.
- Addiction.
- Economic circumstances.
A prisoner may learn to behave appropriately in order to earn tokens without changing the underlying factors that contributed to their offending.
Therefore, token economies may work best when combined with other forms of rehabilitation.
Ethical Issues
Token economies can create ethical problems because prisoners already have limited freedom.
If important privileges are controlled through token systems, offenders may feel pressured to participate.
This raises concerns about coercion.
There is also a danger that basic rights could become confused with rewards.
For example, food, healthcare, safety and essential living conditions should never depend on earning tokens.
Therefore, programmes need clear ethical safeguards.
Staff Consistency
Token economies depend heavily on staff applying rules consistently.
Different officers may interpret behaviour differently.
For example, one officer may consider an interaction respectful while another may not.
This creates problems with reliability.
Inconsistent reinforcement may also weaken learning because offenders cannot clearly predict which behaviour will receive reinforcement.
Therefore, staff training and clear operational definitions are essential.
Cultural Bias
Judgements about what counts as appropriate behaviour can also be influenced by cultural expectations.
For example, communication style, eye contact and expressions of emotion can differ between cultural groups.
If staff use culturally narrow definitions of respectful or cooperative behaviour, some prisoners may be unfairly disadvantaged.
Therefore, token economies need carefully defined and culturally sensitive behavioural criteria.
Sample Bias
Classic research often relied heavily on male offenders.
Hobbs and Holt studied 125 adolescent males in an American correctional institution.
Other early correctional behaviour modification research also frequently involved male American prisoners.
Therefore, findings may not generalise equally well to:
- Female offenders.
- Older offenders.
- Different cultures.
- Different prison systems.
- Modern UK offenders.
This limits the population validity of the evidence.
Temporal Validity
Much of the classic research into token economies was conducted during the 1960s and 1970s.
Prison systems have changed considerably since then.
Modern prisons operate under different:
- Ethical standards.
- Rehabilitation programmes.
- Staff practices.
- Legal frameworks.
- Institutional cultures.
Therefore, findings from older research may have limited temporal validity.
Modern evidence suggests that reward systems can still influence behaviour, but their effectiveness depends heavily on how programmes are implemented.
Summary
- Behaviour modification uses operant conditioning to increase desirable behaviour and reduce undesirable behaviour in custody.
- A token economy gives secondary reinforcers for target behaviours, which offenders can exchange for valued back-up reinforcers.
- Positive reinforcement, immediate rewards, consistency and shaping can increase behaviours such as rule-following and participation in education.
- Hobbs and Holt (1976) found improved appropriate behaviour among 125 adolescent males after token reinforcement was introduced.
- Token economies can improve short-term institutional behaviour, but ethical concerns, staff inconsistency and problems with generalisation limit claims about long-term rehabilitation.
Quick Questions
What learning principle provides the main basis for token economies?
Operant conditioning.
Why is a token described as a secondary reinforcer?
A token is a secondary reinforcer because it gains value through association with a reward or privilege that it can later be exchanged for.
Give two examples of target behaviours that could be reinforced in custody.
Examples include following rules, completing work, attending education, maintaining hygiene, cooperating with staff or avoiding aggression.
What did Hobbs and Holt (1976) find after introducing token reinforcement?
Appropriate behaviour increased after the token economy was introduced into the intervention cottages.
Why might behaviour learned through a token economy fail to continue after release?
Tokens and immediate institutional reinforcement are usually unavailable after release, so behaviour may fail to generalise to the community.
Key Terms
Behaviour modification – The systematic use of learning principles to increase desirable behaviour and reduce undesirable behaviour.
Behaviourism – An approach which argues that behaviour is learned from the environment and should be studied through observable actions.
Operant conditioning – Learning through the consequences of behaviour. Behaviour followed by reinforcement becomes more likely to occur again.
Positive reinforcement – Adding a desirable consequence after a behaviour, which increases the likelihood of that behaviour being repeated.
Negative reinforcement – Removing an unpleasant condition after a behaviour, which increases the likelihood of that behaviour being repeated. It is not the same as punishment.
Punishment – A consequence designed to reduce the likelihood that a behaviour will occur again.
Token economy – A behaviour modification system in which offenders earn tokens for target behaviours and exchange them for rewards or privileges.
Secondary reinforcer – A learned reinforcer that gains value through association with another reward. Tokens are secondary reinforcers because they can be exchanged for valued outcomes.
Back-up reinforcer – The reward or privilege for which tokens can be exchanged.
Target behaviour – A specific and observable behaviour that a behaviour modification programme aims to increase or decrease.
Shaping – Reinforcing small steps towards a desired final behaviour.
Response cost – Removing tokens or privileges following undesirable behaviour.
Generalisation – The transfer of learned behaviour from the situation in which it was learned to other situations.
Recidivism – Reoffending after an offender has already been convicted or punished for an offence.