Counterconditioning

1. Introduction

Version followed: Mary Cover Jones's two 1924 reports ("A laboratory study of fear: the case of Peter" and "The elimination of children's fears") for direct conditioning and social imitation; Joseph Wolpe's reciprocal-inhibition procedures (Wolpe 1958; Wolpe 1990, 4th ed.) as summarised by the Athabasca University Wolpe archive and the SAGE/BJPsych secondary sources for systematic desensitisation and the other anxiety-inhibiting responses; Lazarus & Abramovitz (1962) for emotive imagery; Cautela's covert sensitisation (Cautela & Kearney 1986) for the aversive branch. Wolpe's 1958 and 1990 books are not drawn on directly [uncertain where details depend on them].

Counterconditioning is a learning principle and a family of procedures, not a single branded therapy. The idea: a cue that now triggers an unwanted conditioned response (usually fear) is repeatedly paired with a stimulus or response that is incompatible with it (eating, play, relaxation, assertion, pleasant emotion), at an intensity low enough that the incompatible response wins. Each such pairing weakens the cue-fear link and attaches a new response to the cue. Wolpe called this reciprocal inhibition: an anxiety habit is weakened when a response incompatible with anxiety is evoked in the presence of the anxiety cue [source: Athabasca Wolpe archive; Carona et al. 2023]. The aversive branch runs the same logic in reverse: an unwanted appetitive cue (alcohol, a problem behaviour) is paired with an aversive response so the cue loses its pull [source: Encyclopedia.com, Covert sensitization]. Current learning theory treats the old association as inhibited rather than erased, so relapse prevention matters [source: Wikipedia, Exposure therapy; Hendrikx et al. 2021].

2. Framework

There is no single manualised phase model; the phases below are the common structure of Jones's in vivo method and Wolpe's systematic desensitisation (SD), which share the same logic [synthesis of sources below].

Phase A: Assessment and target definition (1–2 sessions). Goals: identify the feared cue(s) and their generalisation (Peter's rabbit fear had spread to fur, cotton, feathers); find the strongest-reacting cue; take a history (onset, what worsens or eases it); establish a baseline with a graded behavioural ladder (Jones) or SUDs ratings and questionnaires such as the Fear Survey Schedule (Wolpe). Move on when the target cue, its generalised cues and a measurable baseline are agreed. [source: Jones 1924a; Carona et al. 2023; Athabasca concept page]

Phase B: Choose and build the competing response (Wolpe: about 6 relaxation lessons, roughly a third of each session; in Jones's in vivo method, no training phase: food, play or peers are used directly). Goals: select a response the client can produce reliably and that is stronger than the lowest-level fear: eating or play (young children), relaxation (specific phobias), assertion (interpersonal anxiety), sexual arousal (sexual anxiety), pleasant or heroic imagery (children, emotive imagery). Move on when the client can produce the response on cue (e.g. reaches deep relaxation from a brief cue, with twice-daily home practice of 10–15 minutes). [source: Athabasca concept page; Carona et al. 2023]

Phase C: Hierarchy construction (often overlapping Phase B). Goals: list situations by theme and rank them by the client's own ratings, not by logic; adjacent items roughly 5–10 SUDs apart; typically 10–20 or 10–30 items per hierarchy; several hierarchies if several fears. Jones used a 17-step tolerance ladder from "caged rabbit anywhere in the room causes fear" up to "lets the rabbit nibble his fingers". Move on when the lowest item is clearly manageable while the competing response is active. [source: Athabasca concept page; Jones 1924a]

Phase D: Counterconditioning proper (imaginal and/or in vivo). Goals: pair each hierarchy step with the competing response, ascending only when the current step no longer disrupts it. Format: Jones treated Peter daily or twice daily for about two months, in brief sessions at a meal or play period; Wolpe's desensitisation segment is 15–30 minutes, starting with 8–10 scene presentations and rising to 30–50, up to four hierarchies in one session. Total sessions in SD reports range from one to about 100. Moving criterion: in SD, a zero SUD on repeated presentation (usually 3–4, sometimes 10+); in Jones's method, the child keeps eating or playing with the cue at the current distance. [source: Jones 1924b; Athabasca concept page]

Phase E: Generalisation and maintenance. Goals: test gains in real settings, add in vivo steps if imaginal gains do not transfer, cover generalised cues (Jones checked fur, cotton, feathers, a mouse, worms), and plan for setbacks. End when the target and the generalised cues are tolerated in real life and the client has a relapse plan. [source: Athabasca concept page; Jones 1924a] Booster sessions over subsequent months are standard in covert sensitisation [source: Encyclopedia.com]; for fear work they are [synthesis].

Format. Individual is standard; Jones worked with young children in an institutional setting and used small peer groups for social imitation. Parents or carers can run home practice for children [synthesis].

3. Guiding principles

Keep the competing response stronger than the fear. Present the cue only at an intensity at which eating, play or relaxation continues; if it stops, you have gone too far. [source: Jones 1924b]
Distance and intensity are the dials. Jones measured fear by the distance at which the child could keep eating; move the cue closer only as tolerance grows. [source: Jones 1924b]
Pace too fast and you condition the wrong way. Jones warned the method "requires delicate handling": a badly paced pairing can attach fear to the food instead. [source: Jones 1924b]
Use the client's own ratings. Hierarchy order follows the client's SUDs even when it looks illogical. [source: Athabasca concept page]
Small steps, but not trivially small. Items about 5–10 SUDs apart; too far apart stalls progress, and too close wastes time. [source: Athabasca concept page]
Ascend only on a clear criterion. Zero SUD, or uninterrupted competing response, before the next step. [source: Athabasca concept page; Jones 1924b]
Match the competing response to the fear. Relaxation for specific phobias, assertion for interpersonal anxiety, sexual responses for sexual anxiety, food or play for young children. Relaxation is not indispensable. [source: Carona et al. 2023; Athabasca concept page]
Avoidance is not treatment. Jones found shielding the child from the cue ("disuse") "an unsafe method": the fear returned. [source: Jones 1924b]
Talk alone does not change the response. Reassurance and storybook appeals did not change a child's reaction to the real animal. [source: Jones 1924b]
Never shame or tease. Ridicule failed and in one case strengthened the fear. [source: Jones 1924b]
Use social influence deliberately. Calm, admired peers or adults near the cue help; frightened adults near the cue harm (Peter's relapse followed a dog jumping at him and his frightened nurse). [source: Jones 1924a, 1924b]
Combine methods. Jones found direct conditioning and social imitation reliably effective, and the other methods useful only alongside them; later reviewers also advised combining techniques. [source: Jones 1924b; Rachman & Costello 1961 abstract]
Generalise deliberately. Treat the strongest cue, then check and treat the transferred cues and real settings. [source: Jones 1924a; Athabasca concept page]
End sessions on success. Finish on a fully desensitised step or clear progress. [source: Athabasca concept page]
Track everything. Log SUDs or ladder steps every session; plateaus and relapses are normal and show up only in records. [source: Athabasca concept page; Jones 1924a]
Consent and choice. Explain the rationale, get consent (and a child's assent), and agree that the client can stop any step. Informed consent is described as essential in covert sensitisation. [source: Encyclopedia.com; extension to all procedures is synthesis]

4. Intervention inventory

Items 1–6 are Jones's in vivo methods (1924); items 7–15 are Wolpe-lineage reciprocal-inhibition procedures and their child variants; items 16–18 are the aversive branch, included for completeness with explicit limits. All example language is illustrative unless quoted.

1. Behavioural tolerance ladder (assessment and progress measure)

Purpose / when: At intake and every session, to measure fear as observable approach behaviour, especially with children who cannot give ratings. [source: Jones 1924a]
Steps: (1) Run brief test exposures to the feared cue and its likely generalised cues; note which provokes the strongest reaction and make it the first target. (2) Write a ladder of approach behaviours from "cue anywhere in the room provokes fear" through decreasing distances, cue free in the room, touching it while you hold it, touching it free, holding it, staying alone with it, to affectionate handling. Jones's ladder had 17 steps (A–Q). (3) Record the highest step reached each session, with date and any events. (4) Optionally have a colleague rank the steps blind to check the order is sensible (Jones had six raters; average agreement .70). [source: Jones 1924a]
Example language (illustrative): "Let's see how close Mr Rabbit can be while you keep playing. You tell me if he's too close."
Watch for: Steps are not equal intervals; expect plateaus, sudden gains and relapses. Jones's chart also hid unequal gaps between sessions, so record dates. Stop physiological measures if the measuring itself frightens the child (Jones stopped blood-pressure readings for this reason). [source: Jones 1924a]

2. Direct conditioning with food or another appetitive response (in vivo counterconditioning)

Purpose / when: The core Jones procedure; for specific fears (animals, objects) in young children who cannot use relaxation or imagery. One of the two methods Jones judged unqualified successes. [source: Jones 1924b]
Steps: (1) Schedule the session when the child is hungry or at a natural snack or meal time; seat the child comfortably (Jones used a high chair at mid-morning lunch) with a food they like. (2) Introduce the cue (Jones used the rabbit in a wire cage) at a distance where the child keeps eating calmly. (3) If eating stops or the child shows fear, move the cue back until eating resumes. (4) Over successive pairings, move the cue closer: nearer the chair, on the table, within touching distance. (5) Advance one ladder step at a time; record the distance at which eating continues as the fear measure. (6) Repeat daily or more often (Peter: daily or twice daily for about two months). (7) Move to touching and handling once the cue is tolerated at close range. [source: Jones 1924a, 1924b]
Modern adaptation [synthesis]: Where food is unsuitable (feeding difficulties, eating-disorder risk, allergies, parental preference), use another strong positive activity the child engages in, such as favourite play, music or a game, with the same distance rule. Do not use food with clients who have disordered eating.
Example language (illustrative): "You keep eating your snack. Puppy is staying way over by the door." Later: "You're still munching — shall we let puppy come one square closer?"
Watch for: Jones's central caution: the method "requires delicate handling"; a pairing that is too intense can attach fear to the food instead. Signs it is working: eating continues at closer distances; spontaneous looking, then approach. Signs of overload: stopping eating, freezing, crying, pushing food away → increase distance at once and end on an easy step. [source: Jones 1924b]

3. Social imitation (peer modelling)

Purpose / when: The second method Jones judged unqualified successes; use alongside direct conditioning, especially with children who are influenced by admired peers. [source: Jones 1924b]
Steps: (1) Recruit one to three children (or siblings, or a calm adult the child likes) who are unafraid of the cue and generally well adjusted. (2) Hold a play period with the cue present for part of the time. (3) Let the models handle and talk about the cue naturally; do not pressure the target child. (4) Allow the target child to approach in their own time; praise approach. (5) Alternate group sessions with solo sessions so you can see the child's own level. [source: Jones 1924a, 1924b]
Example language (illustrative): To the model: "Would you like to show everyone how you give the bunny a carrot?" To the target child: "You can watch from wherever you like."
Watch for: Fear spreads socially too: a frightened peer or adult near the cue can undo progress (Peter's relapse; Vincent acquired his fear from a crying peer). Choose models who are calm and avoid group teasing. Working signs: the child edges closer after watching, imitates the model's handling, laughs. [source: Jones 1924a, 1924b]

4. Distraction (adjunct only)

Purpose / when: Jones found a substitute activity near the cue gave "fair results" and helped one child approach, but effects were temporary and required an adult constantly present; use only as a bridge into direct conditioning. [source: Jones 1924b]
Steps: (1) Place an absorbing activity (Jones used crayons) near the cue at a tolerable distance. (2) Let the child engage with it. (3) Gradually bring the activity and cue closer together. (4) Fade the distractor as the child begins to attend to the cue without fear. [source: Jones 1924b; step 4 is synthesis]
Example language (illustrative): "Let's draw over here at this table. Froggy is in his tank by the window."
Watch for: Modern exposure guidance generally discourages distraction that stops the client noticing the cue, because it can limit new learning [synthesis; uncertain]. Use it to get started, then make sure the child is attending to the cue while calm.

5. Methods Jones tested and found unreliable alone (what not to rely on)

Purpose: To prevent common well-meant errors. [source: Jones 1924b]
Disuse: Shielding the child from the cue for weeks; fears returned or persisted. Do not advise avoidance as treatment.
Verbal appeal: Talking about the cue and linking it to pleasant stories or toy versions; did not change the reaction to the real cue. Use only as preparation.
Negative adaptation: Repeated, monotonous presentation without anything positive; worked in one case but Jones noted it aims at indifference rather than acceptance and often intensified fear. If you use plain repeated exposure, use modern graded-exposure safeguards.
Repression: Ridicule and teasing; failed and sometimes strengthened fear. Never use.
Watch for: Parents and teachers often use these methods spontaneously; ask about them in assessment and coach alternatives [synthesis].

6. Combined in vivo package for children (Jones's Peter protocol, as a whole)

Purpose / when: Jones combined ladder, direct conditioning, peer modelling and positive adult presence; use this combination as the default for a young child's specific fear. [source: Jones 1924a]
Steps: Daily or twice-daily brief sessions; group play with fearless peers with the cue present; separate feeding pairings starting from the second ladder step; record ladder step each session; after the main cue is mastered, test and treat generalised cues (Peter's cotton, fur, feathers; then novel animals). [source: Jones 1924a]
Example language (illustrative): To parent: "Each day we'll do two short goes: one with his friends and one at snack time. Our only rule is that he keeps enjoying himself."
Watch for: Illness, a separation or a frightening incident can return fear close to baseline (Peter after two months in hospital and the dog incident); plan for re-starting lower on the ladder. Jones also cautioned that a difficult home environment limits predictions. [source: Jones 1924a]

7. SUDs scale

Purpose / when: Before hierarchy building and throughout SD; the client's 0–100 rating of discomfort. [source: Athabasca concept page]
Steps: (1) Anchor 0 as the calmest the client can imagine being and 100 as the worst anxiety they have ever felt. (2) Have them rate a few known situations to calibrate. (3) Use it to rank hierarchy items and after every scene presentation. (4) Do not compare one client's numbers with another's. [source: Athabasca concept page]
Example language (illustrative): "On a 0 to 100 scale, where 0 is completely calm and 100 is the most anxious you've ever been, where would you be standing at the top of the escalator?"
Watch for: Clients who rate everything high or low; recalibrate with anchors. SUDs correlate with physiological measures such as heart rate, but self-report is the working measure. [source: Athabasca concept page]

8. Hierarchy construction

Purpose / when: After assessment, before desensitisation. [source: Athabasca concept page; Carona et al. 2023]
Steps: (1) Gather situations from interview and fear questionnaires. (2) Group them into themes (one hierarchy per theme). (3) For each item write a specific, concrete scene (who, where, what, how close, how long). (4) Have the client rate each item in SUDs and order them. (5) Check gaps: adjacent items about 5–10 SUDs apart; insert intermediate items where gaps are larger. (6) Aim for roughly 10–30 items. (7) Include situations not yet experienced, rated by expected anxiety. (8) Revise during treatment, especially by adding easier items. [source: Athabasca concept page]
Example language (illustrative): "Let's make this scene more exact: you're in the car park, the dentist's door is about 20 metres away, it's ten minutes before your appointment. What number does that get?"
Watch for: Items that are vague (hard to imagine consistently), mixed themes, or a lowest item that already exceeds about 20–25 SUDs [threshold is synthesis]. The order must follow the client's ratings, not yours.

9. Relaxation training as the competing response (Wolpe's abbreviated Jacobson method)

Purpose / when: Wolpe's default anxiety-inhibiting response for SD; for adolescents and adults with specific phobias and situational anxieties. [source: Athabasca concept page; EBSCO]
Steps: (1) Explain that tension and anxiety are hard to sustain in a deeply relaxed body. (2) Over about six lessons (roughly a third of each session), teach alternately tensing and releasing muscle groups: arms, face, jaw, eyes, neck and shoulders, back, abdomen and chest, legs. (3) Direct attention to the contrast between tension and release. (4) Set home practice twice daily, 10–15 minutes. (5) Progressively shorten the routine until the client reaches deep relaxation from a brief cue, without tensing first. (6) Check depth with a SUDs rating (aim near 0) before starting scenes. [source: Athabasca concept page; EBSCO] Alternatives named: autogenic training, meditation, yoga, EMG biofeedback. [source: Athabasca concept page]
Example language (illustrative): "Make a fist with your right hand… notice the tightness… and let it go. Notice the difference as the tension drains out."
Watch for: Some clients feel more anxious as they relax (relaxation-induced anxiety) [synthesis]; switch to another competing response. Relaxation is not indispensable; later research found graded imaginal exposure worked about as well without it. [source: Carona et al. 2023; Athabasca concept page]

10. Systematic desensitisation (imaginal)

Purpose / when: Wolpe's central procedure for anxiety disproportionate to the situation, when in vivo steps are impractical, or as preparation for in vivo work. [source: Athabasca concept page; Carona et al. 2023]
Pre-start safety and consent [synthesis]: Explain the procedure, that the client controls stopping, and that anxiety will be kept low. Screen for conditions in which deliberate arousal or imagery needs medical or specialist input (significant cardiac or respiratory disease, pregnancy with medical complications, seizure disorders), active psychosis, acute suicidality, intoxication, or a dissociative response to imagery; address these first or adapt.
Steps: (1) Induce relaxation (or the chosen competing response). (2) Control scene: present a neutral scene unrelated to the fear to check that the client can visualise and can stop imagining on request. (3) Agree a signal: the client lifts a finger slightly when the scene is clear. (4) Present the lowest hierarchy item; once signalled, hold it about 5–7 seconds, then ask the client to stop the scene. (5) Get a SUD rating. (6) Give 10–30 seconds of renewed relaxation. (7) Re-present the same item until SUD reaches 0 (usually 3–4 presentations; sometimes 10 or more). (8) Move to the next item only after a zero rating. (9) If the client signals marked anxiety mid-scene, stop the scene immediately and return to relaxation. (10) End the session on a fully desensitised item or clear progress; log every rating. (11) Next session, resume at the last unresolved item (or, in some authors' practice, at the last fully desensitised item). [source: Athabasca concept page; EBSCO]
If an item stalls: shorten presentation time; spend longer (about 20 seconds) on the item below; insert an intermediate item. [source: Athabasca concept page]
Example language (illustrative): "Stay as relaxed as you are now. When I describe the scene, picture it as clearly as you can, and lift your finger when you have it… You're standing at the bottom of the escalator, watching people step on… Now stop the scene and let it go. Just relax. What number?"
Watch for: Rising SUDs across repetitions (step too large); inability to visualise (switch to in vivo or more concrete scripts); avoidance within the image (changing details); falling asleep. Working signs: SUDs fall across repetitions and sessions, and real-life approach starts spontaneously. Imaginal gains may not transfer; test in vivo. [source: Athabasca concept page; EBSCO; first sentence partly synthesis]

11. In vivo desensitisation

Purpose / when: To transfer imaginal gains to real life, or as the main method when imagery is weak or the cue can be graded safely in reality. [source: Athabasca concept page; SAGE (McGlynn 2005); EBSCO]
Steps: (1) Build a real-life hierarchy (or reuse the imaginal one). (2) Before each step, establish the competing response (relaxation, calm breathing, the presence of a trusted person, a pleasant activity). (3) Enter the lowest real step for as long as the competing response holds. (4) Retreat a step if anxiety climbs past an agreed ceiling; re-relax; try again. (5) Repeat until the step is easy, then ascend. (6) Assign homework steps between sessions. Virtual-reality or computer-simulated steps can replace in vivo where real exposure is impractical or unsafe. [source: EBSCO; Athabasca concept page; steps 2–4 partly synthesis]
Example language (illustrative): "We'll stand at the corner where you can just see the dog park. Let your shoulders drop. Tell me your number every minute or so."
Watch for: Real cues are less controllable than imagined ones: plan exits, use trained animals and safe settings, and check allergies and bite or scratch risk with animal phobias (Peter's scratch caused a sharp drop) [source: Jones 1924a; safety steps are synthesis]. Respect withdrawal of consent at any step.

12. Assertive responses (assertion as counterconditioner)

Purpose / when: Wolpe's first reciprocal inhibitor; for anxiety evoked by other people (difficulty refusing, expressing irritation or affection). Not helpful for anxiety from non-personal cues. [source: Athabasca Wolpe biography; Carona et al. 2023]
Steps [uncertain: detailed steps from Wolpe 1990 not drawn on; standard behaviour-therapy practice]: (1) Identify situations where the client inhibits legitimate expression and feels anxious. (2) Explain that expressing appropriate feelings tends to inhibit the anxiety. (3) Grade situations from least to most threatening. (4) Rehearse each in session through role play, with the therapist playing the other person, coaching voice, posture and wording; reverse roles to model. (5) Assign real-life assertions starting low, where a successful outcome is likely. (6) Review outcomes and ascend.
Example language (illustrative): "Let's practise the first one: your colleague asks you to stay late again. Say no in one sentence, look at me, and keep your voice level."
Watch for: Aggression in place of assertion; homework set where the social consequences could be harmful (e.g. unsafe relationships) [synthesis]. Working signs: less anticipatory anxiety, spontaneous assertion.

13. Sexual responses (for sexual anxiety)

Purpose / when: Wolpe used sexual arousal as the competing response for anxiety that interferes with sexual functioning. [source: Athabasca Wolpe biography; Carona et al. 2023; SAGE (McGlynn 2005)]
Steps [uncertain: detailed instructions in Wolpe 1958/1990 not drawn on; outline only]: (1) Rule out medical causes and confirm a willing, informed partner. (2) Explain that anxiety blocks arousal and that arousal can in turn inhibit low-level anxiety. (3) Ask the couple to engage only in physical intimacy at levels where the client feels pleasurable arousal and no more than slight anxiety, stopping or stepping back if anxiety rises. (4) Remove performance demands (e.g. agree not to attempt intercourse until earlier levels are anxiety-free). (5) Ascend step by step across sessions. The structure overlaps with later sensate-focus methods [synthesis].
Example language (illustrative): "For the next two weeks, the only goal is enjoyment at the level you're comfortable with. If anxiety rises, slow down or go back a step."
Watch for: Partner pressure, relationship conflict, trauma history (which may need trauma-focused treatment first) [synthesis].

14. Emotive imagery (children)

Purpose / when: A child variant of SD (Lazarus & Abramovitz 1962) for children who cannot do muscle relaxation: positive emotions (excitement, pride, affection, mirth) aroused by a story about the child's hero serve as the competing response. Studied for fears such as darkness (six weekly sessions in Cornwall et al. 1996). [source: Cornwall et al. 1996 abstract; informahealthcare abstract; procedure from Lazarus & Abramovitz 1962 — uncertain]
Steps [uncertain: as commonly described]: (1) Build a graded fear hierarchy with the child and parent. (2) Find out the child's heroes and favourite fantasies. (3) Ask the child to imagine a story close to everyday life in which the hero appears and enlists the child. (4) When the child is clearly engaged and excited, weave in the lowest hierarchy item as part of the story. (5) If the child shows anxiety, take the item out and return to the story. (6) Repeat and ascend; follow with in vivo practice and praise for real-life brave steps.
Example language (illustrative): "Your favourite superhero needs a secret agent to guard the house tonight. He's chosen you. You're in the hallway with your special torch, and the hall light goes off… how are you doing, agent?"
Watch for: Story too exciting to include the cue; the hero doing the brave act instead of the child; the child going along to please you. Check real-life behaviour, not only the story.

15. Other competing responses (pleasant imagery, humour, activity, breathing)

Purpose / when: When relaxation does not suit. Named alternatives include pleasant thoughts, humour and laughter, physical activity, eating, assertion and sexual arousal; Wolpe matched breathing responses to generalised anxiety. [source: Athabasca concept page; EBSCO; Carona et al. 2023]
Steps: (1) Pick a response the client produces reliably and enjoys. (2) Test that it holds against the lowest item. (3) Pair it with hierarchy items exactly as in items 10–11. Experimental variant: imagining a pleasant outcome (e.g. a pleasant sound) right after each feared-cue presentation during exposure with response prevention. [source: Hendrikx et al. 2021]
Example language (illustrative): "After each picture of the spider, bring back the sound of your favourite song as vividly as you can."
Watch for: In the Hendrikx et al. study, positive imagery reduced distress during exposure but did not improve avoidance or fear outcomes, and the authors advised caution about clinical use; treat it as a tolerability aid, not an enhancer [source: Hendrikx et al. 2021]. Wolpe's breathing response used carbon-dioxide inhalation [uncertain]; this is not current practice. If you use slow calm breathing instead, that is a modern substitute [synthesis].

16. Covert sensitisation (aversive counterconditioning in imagery)

Purpose / when: Cautela's procedure pairs an unwanted approach behaviour with imagined aversive consequences and pairs refusal with imagined relief. Described uses: alcohol, drug and tobacco use, gambling, overeating, and harmful sexual behaviours such as offending-related arousal. It uses imagery only, unlike physical aversion therapy. [source: Encyclopedia.com; SAGE (Emmelkamp & Kamphuis 2005)]
Never use to change sexual orientation or gender identity; such conversion practices are harmful and widely prohibited or condemned by professional bodies [synthesis from general professional consensus]. Use only for behaviour the client wants to stop, with full consent.
Steps: (1) Assess frequency, severity and settings of the behaviour. (2) Explain the rationale; obtain informed consent, since the client will deliberately imagine distressing images. (3) With the client, list personally meaningful aversive images (e.g. nausea, social embarrassment) and choose the strongest the client is willing to use. (4) Teach relaxation to support vivid imagery. (5) Aversive scene: the client imagines approaching the usual setting and, just before carrying out the behaviour, vividly experiences the aversive consequence. (6) Escape/relief scene: the client imagines turning away and feeling relief as the unpleasant sensation fades. (7) Self-control scene: the client imagines refusing before the behaviour starts and feeling good. (8) Alternate scenes in session; the therapist narrates if the client struggles to imagine. (9) Homework: the client practises the scenes (often 10–20 per practice) and later uses the imagery when tempted in real life. (10) Suggest an incompatible alternative behaviour. (11) Typically 5–20 office sessions, then booster sessions over 6–12 months plus relapse prevention. [source: Encyclopedia.com]
Example language (illustrative): "You're walking up to the bar… you reach for the glass… and your stomach turns over… you put it down and step outside into the fresh air, and the queasiness eases. You feel relief and pride."
Watch for: Images too weak to produce any aversion, or so intense the client avoids sessions; generalised low mood or disgust spreading to neutral cues [synthesis]. Keep it within a broader treatment (e.g. relapse prevention, motivational work) [synthesis]. Sources describe it as comparatively low-risk and list no specific contraindications. [source: Encyclopedia.com]

17. Anxiety-relief conditioning (historical; not recommended)

What it was: A Wolpe procedure in which the ending of an unpleasant stimulus was paired with a calming word, which the client could later say to themselves when anxious. [uncertain: from Wolpe 1958/1990; listed among Wolpe's methods in the Athabasca biography as "aversion relief"]
Status: It relied on mild electric shock. It is not current practice and is described here only for completeness; no operational steps are given. Use modern alternatives (cue-controlled relaxation, items 9–11) [synthesis].

18. Physical aversion therapy and aversion relief (historical; not recommended)

What it was: Pairing an undesired behaviour or stimulus with a physical aversive (electric shock, nausea-inducing chemicals, bitter tastes), sometimes with relief when the person switches to a desired stimulus or stops the behaviour. [source: SAGE (Emmelkamp & Kamphuis 2005)]
Status: Chemical and electrical aversion carry greater risks than imagery methods [source: Encyclopedia.com]. It was historically used to try to change sexual orientation, now recognised as harmful [synthesis]. Do not use; where an aversive approach is clinically justified and wanted by the client, covert sensitisation (item 16) within a broader treatment is the lower-risk option [synthesis].

5. Handling problems

Most of this section is practical guidance derived from the sources' principles; source-based points are tagged.

The competing response collapses (child stops eating, client cannot stay relaxed). Move the cue farther away or go back one item at once; re-establish the competing response; end on an easy success. Jones's distance rule is the model. [source: Jones 1924b; Athabasca concept page]
Stuck on one item. Shorten exposure time; spend about 20 seconds on the item below; insert an intermediate item; re-rate the whole hierarchy. [source: Athabasca concept page] Check whether the item has a hidden element (e.g. being trapped, being judged) that needs its own hierarchy [synthesis].
Overwhelm (marked fear, crying, panic). Stop the scene or retreat from the cue immediately; return to the competing response; do not end the session at the peak; next time start two steps lower [source: Athabasca concept page for stopping; rest synthesis]. With a child, a calm adult nearby matters: Jones linked Peter's relapse partly to a frightened adult. [source: Jones 1924a]
Fear attaches to the reward (refuses the snack, dislikes the relaxation room). Jones's warned-of reversal: pause the pairing, rebuild enjoyment of the reward without the cue, then restart at a much greater distance; consider switching to a different reward. [source: Jones 1924b for the risk; responses are synthesis]
Relapse after illness, a separation or a new frightening event. Expect it (Peter returned near baseline after hospital and the dog incident). Re-assess, restart lower on the ladder; recovery is usually quicker the second time [source: Jones 1924a for the relapse; the faster-recovery expectation is synthesis].
Small setbacks (a scratch, a startle). Jones saw a sharp drop after a scratch, then quick recovery: continue at a slightly lower step rather than stopping. [source: Jones 1924a]
Gains don't transfer from imagination to real life. Add in vivo desensitisation; test real settings. [source: Athabasca concept page]
Cannot visualise. Use the control scene to diagnose; add sensory detail, use photos, video or VR, or switch to in vivo. [source: Athabasca concept page for the control scene; EBSCO for VR; rest synthesis]
Cannot relax or relaxation increases anxiety. Switch to another competing response (activity, humour, pleasant imagery, a trusted person's presence) or to graded exposure without relaxation, which later research found roughly as effective. [source: Athabasca concept page; Carona et al. 2023]
Non-adherence to home practice. Shrink homework to the smallest step the client will do; schedule it to existing routines; have a parent run children's practice; review logs at the start of each session [synthesis].
Resistance or refusal to approach. Never force: forced contact risks strengthening fear [synthesis consistent with Jones's caution]. Re-explain the rationale; let the client set the pace and the next step; use peer or therapist modelling first. [source: Jones 1924a, 1924b for modelling]
Family members teasing, reassuring endlessly or shielding the child. Jones found ridicule, verbal appeal alone and disuse ineffective or harmful. Coach carers to model calm approach, praise small steps and stop teasing. [source: Jones 1924b; coaching is synthesis]
Rupture (client feels pushed or tricked). Name it, own the pacing error, restore control ("you say when we go on"), and step back down the hierarchy [synthesis].
Distraction or safety behaviours used to get through. Allow early, fade later, so the client learns the cue is tolerable while attending to it [synthesis; uncertain].
Multiple fears or a broad anxiety disorder. Build separate hierarchies (up to four can be worked on in one session) and consider a fuller CBT protocol if fears are pervasive. [source: Athabasca concept page; the referral point is synthesis]
Aversive work (item 16) produces distress or demoralisation. Reduce image intensity, increase relief and self-control scenes, and set it inside a broader relapse-prevention plan; stop if the client withdraws consent [synthesis].
Risk. If the client reports suicidal thoughts or self-harm, pause the procedure and complete a standard risk assessment and safety plan, including lethal-means counselling (safe storage or removal) where indicated, before resuming [synthesis; standard practice].

Sources

Sources:

Books and papers cited (Jones 1924a and 1924b are covered via the pages above; the others are not drawn on):

Jones, M. C. (1924a). A laboratory study of fear: The case of Peter. Pedagogical Seminary, 31, 308–315.
Jones, M. C. (1924b). The elimination of children's fears. Journal of Experimental Psychology, 7, 382–390.
Wolpe, J. (1958). Psychotherapy by Reciprocal Inhibition. Stanford University Press.
Wolpe, J. (1990). The Practice of Behavior Therapy (4th ed.). Pergamon.
Lazarus, A. A., & Abramovitz, A. (1962). The use of "emotive imagery" in the treatment of children's phobias. Journal of Mental Science, 108, 191–195.
Cautela, J. R., & Kearney, A. J. (1986). The Covert Conditioning Handbook. Springer.

← Previous · II-006Systematic desensitisation Next · II-008 →In-vivo, imaginal and interoceptive exposure