Table 2.

Predictions/findings of altered reinforcement sensitivity in ADHD, expected impact on child behavior, and recommended practices for inclusion in behavioral parent and teacher training programs

Prediction/Findinga	Strength of evidenceb,c	Theoriesd	Potential impact on child’s behavior	Recommendations for management
Positive reinforcement
Prefer immediate over delayed reward	Very strong	DTD [1] DDT [1] Delay aversion [3] Douglas [1]	Act to avoid or minimize delay Impulsivity	Reinforce immediately where possible Educate caregivers to avoid/reduce experience of delay for children with ADHD Educate caregivers children may act to avoid/reduce delay resulting in impulsive behavior Link rewards to actions before and when delivering reward Teach older children how they can manage delay and self-reward Enhance the saliency of the expected behavior before and during delay-related situations (e.g., clear instructions, pictures of expected behavior), if possible reduce saliency of incidental immediate rewards in the child’s environment (e.g., remove distractors)
Escape/avoid situations that involve delay	Very strong	Delay aversion [2]	Impulsivity
Impaired acquisition of conditioned reinforcers/impaired response to reward-predicting cues			More distractible off-task behavior Poorer stimulus control Poorer maintenance of behavior by delayed or discontinuous reinforcement	Draw child’s attention to rules/expectations of the current context and when these change Increase saliency of cues Scaffold/shape learning of adaptive behavior Highlight the outcome of the child’s behavior to them
Imaging studies	Very strong	DTD [6]
Behavioral studies	None	DDT [5]
Show more variable responding, with only short sequences of behavior reinforced
More variable responding	Very strong	DDT [3]	More variable task performance	Maintain high rates of immediate reinforcement throughout tasks
Short sequences of behavior reinforced	None
No performance deficit under continuous reinforcement  Poorer performance under partial reinforcement	Limited (3) Limited/mixed (4/2)	DTD [3,2] DDT [2]	Learn less/more slowly under partial reinforcement Slower adaption of behavior to new situations/contingencies under partial reinforcement	Maintain high rates of reward during learning of new behaviors/skills (acquisition) Institute individualized stretching the ratio’s (gradually shift from more continuous to partial schedules of reinforcement) to install persistence of learned behavior. In doing so ensure rates of reinforcement are sufficient to maintain interest in a task/activity
Increased vulnerability to possible arousing and distracting effects of reward	Limited/mixed (1/1)	Douglas [2]	Reward may not be as effective as expected	Monitor responses to reward, revise reinforcement schedules as required over time
Increased influence of individual/recent occurrences of reinforcement on behavior	Limited (1 direct) (1 indirect)	DTD [7]	Increased risk of non-target behaviors being reinforced by other events/incidental rewards More off-task behavior due to increased distractibility toward incidental rewards Stronger environmental & weaker internal control of behavior	Educate caregivers of children’s vulnerability to incidental rewards, increasing the probability of non-adaptive behaviors being reinforced Remind children of setting specific rules Increase the saliency of discriminative stimuli to reduce the impact of incidental rewards Reduce distracting/competing recent rewards by increasing environmental and temporal structure
Faster extinction	Limited (1 + 2 indirect)	DTD [4] Non-specific	Less behavioral persistence Learn/demonstrate less new exploratory/adaptive behavior in the context of refraining from formerly rewarded behavior	Use of stretching the ratio’s to install behavioral persistence Don’t assume children will “know what to do” when you stop rewarding a learned behavior. Remind children of the expected adaptive behavior once extinction starts, i.e., when reward for adaptive behavior stops or is reduced in frequency For older children teach them to self-reinforce under conditions of extinction
Differential extinction effects	Limited/mixed (3/2)
Heightened emotional response when anticipated rewards fail to appear	Limited (2)	Douglas [3] Amsel [1]	More frustration in daily life (continuous reinforcement in daily life is rare)	Strive for consistency of rewards as much as possible Educate caregivers about emotional effects of reinforcement manipulations (e.g., response cost/ ignoring/ partial reinforcement) Some children may need to be taught emotion regulation techniques
Reduced persistence of behavior in response to partial reinforcement/punishment/extinction due to prolonged frustration	Limited (1 direct) (1 indirect)	Amsel [2]	Higher levels of frustration in daily life interfere with learning persistent adaptive behavior	Educate parents about frustrative effects of non-reward Teach frustration management techniques to parents/children Stretching the ratio’s to gradually expose child to frustration and install persistent adaptive behavior
Additional relevant findings
Poorer matching of behavior to reinforcement contingencies	Moderate/mixed (5/2)	Non-specific	Poorer/slower adaptation of behavior to changing environmental expectations	Reduce behavior-consequence ambiguity:  Explicitly inform children of expectations/rules ahead of time  Explicitly inform child when rules/ contingencies change Institute overlearning Provide sufficient opportunity to transition when rules/requirements change Reinforce desired behavior going into new situations Educate caregivers of children’s difficulty modulating behavior, especially when/if rules/expectations are not made explicit
Failure to adapt behavior to situational demands (when there is a delay)	Limited (1)	Non-specific (CDL)	Environment does not serve as a cue for adaptive behavior	Increase the saliency of the situation and the situation-response association Use Differential Outcomes, i.e., apply different reinforcers for different adaptive behaviors/skills
Negative punishment
Mild punishment (i.e., response cost) improves on-task behavior and performance	Strong/mixed (11/3)	Non-specific	Reduces undesirable behavior in the short-term, long-term effects unclear Potentially more errors under threat of punishment	Educate caregivers of potential negative side-effects Use cautiously Avoid extended use Monitor the emotional response of children to negative punishment Reward alternative adaptive behaviors as an alternative to use of punishment Some children may benefit from learning emotion regulation techniques
Mild punishment (e.g., ignoring/ failure to deliver rewards) may	Limited	Amsel [1]	Increased emotionality to ignoring/non-delivery of rewards Faster reduction/stopping of learned adaptive behavior when no reward follows
Increase frustration	(2 + 1 indirect)
Reduce behavioral persistence	(1)
Positive punishment
Positive punishment causes unwanted side-effects in ADHD	Limited (3)	Non-specific	More errors on tasks Missed learning opportunities (resulting from efforts to avoid punishment) Lack of persistence in activities perceived as punishing (including tasks experienced as frustrating and/or effortful) Slower responding in punishment situations	Educate caregivers about negative side-effects of punishment Avoid use wherever possible Use all the other recommendations to prevent use of punishment (make rules/ consequences explicit/ provide non-punitive calm reminders/ reinforce adaptive behavior)
Not currently supported
Diminished PREE	None	DTD [5]
Slower extinction	None	DDT [4]
Hyperactivity is due to increased responsiveness to stimulation from anticipatory frustration in partial reinforcement conditions	None	Amsel [3]

^aFindings not linked to a specific theory are presented in italics

^bStrength of evidence based on the literature reviewed in the manuscript: Very strong = multiple empirical studies in support of the prediction/finding, including at least one published meta-analysis; Strong = more than 10 empirical studies available in support of the prediction/finding; Moderate = five to 10 empirical studies available in support of the prediction/finding, number of studies in brackets; Limited = less than five empirical studies available in support of the prediction/finding, number of studies in brackets; Mixed = some empirical studies available whose data counters the prediction/finding, number of studies in brackets (support/counter); None = currently no evidence available to support the prediction

^cStrength of evidence here refers to the number of available studies, not the quality of the studies or sample sizes

^dNumbers in square brackets reflect prediction numbers in the text; Non-specific indicates empirical findings not linked to specific theories