A cat that hides from the vacuum 6 times and still hides has learned nothing. A cat that met the carrier once and screams at it for the next 3 years has learned something very well. Both are learning, and both follow rules you can predict. Habituation drops a response after 6 to 10 boring repeats. Sensitization raises it after 1 bad event, and 1 bad event usually outranks 10 calm ones. Which rule is running decides whether 4 weeks of work pays off or makes the cat worse.
Two rules, one nervous system
Habituation is a response shrinking because the thing keeps happening and nothing bad follows. The 7 a.m. can opener stops waking the cat by week 2, not because the cat got brave but because the event stopped carrying information. The conditions are narrow: the stimulus repeats, it stays under the cat's reaction threshold, and the cat can walk away. Remove any 1 of those 3 and the same repetition produces the opposite result. Sensitization is the mirror image. A single unpleasant or startling event lowers the threshold, so the next exposure needs less input to produce a bigger response. It is faster than habituation, it lasts longer, and it spreads: a cat startled by a falling pan at 2 metres may avoid the whole kitchen for 3 weeks.
Rule 1: habituation needs 6 to 10 quiet repeats
Six to 10 exposures is the working range for a mild trigger such as a doorbell, a carrier parked in the hallway, or a new food bowl. One session a day, 3 to 5 minutes a session, 24 hours apart, intensity held at 2 out of 10: ears forward, pupils normal, tail relaxed, still eating. Six sessions at 2 out of 10 do more than 2 sessions at 6. If the cat stops taking a treat mid-session, the intensity is too high, so halve it and step back 1 stage. Once the cat is under threshold, expect the response to fall by about half every 2 sessions. If nothing moves after 4 sessions, the trigger is probably not the problem; pain, hearing loss, or a second trigger in the same room usually is.
Rule 2: sensitization needs 1 event and 10 seconds
Ten seconds is enough. A cat held down for 90 seconds for a nail trim can come out hand-shy for 6 months, and repeating those 90 seconds 4 more times does not make it easier, it makes the cat faster to escalate. The marker is direction, not intensity: if the response grows over 3 repeats, the cat is sensitizing, and more practice will deepen it. Recovery is not passive. Give 48 to 72 hours with zero exposure before the next attempt, then restart 2 stages below where the incident happened. In practice that means 1 step back for every 1 step forward, which is why 10 short sessions at 2 out of 10 beat 1 long session at 7.
The 4-question test after every session
Ask 4 things and write them down, because memory is unreliable at 24 hours. First, did the response get smaller or bigger than last time? Second, how many seconds until the cat went back to eating or grooming: under 60 is fine, over 300 means the session was too hard. Third, did the cat approach the trigger within 2 minutes after the session ended? Fourth, did the same response show up in a second room? Shrink, fast recovery, approach and no spread means habituation. Growth, slow recovery, avoidance and spread means sensitization, and the plan should move backwards, not forwards. Four questions, 30 seconds of notes, and the trend across 3 sessions is visible.
Spacing decides which rule wins
Three gaps matter. Under 10 minutes, repeats fuse into 1 continuous event and the cat sensitizes instead of habituating. At 24 hours the memory consolidates and the response is measurably lower on the second day. At 48 hours a strong fear has a better chance of settling, and 7 days is enough for a mild one and far cheaper to keep up for 3 months. Ten short sessions also beat 1 long one by a wide margin: 10 minutes of flooding is 1 sensitizing event, not 10 exposures, and it can set the plan back 2 to 4 weeks.
Where the rules break: pain, age and hearing
Both rules assume a healthy cat. A 12-year-old cat that suddenly stops tolerating being picked up is more likely to have dental pain or arthritis than a training problem, and no amount of habituation fixes a sore hip. Hyperthyroidism raises baseline arousal, and a cat with a resting rate over 220 beats per minute will not settle for a 5-minute session. Deaf cats habituate to visual triggers only, so a doorbell plan does nothing for them. Cats with reduced vision startle more easily and sensitize in 1 exposure instead of 6. Breed and temperament shift the starting point rather than the rule itself, which is what the short quiz at breed quiz is for.
A 10-day habituation plan
Pick 1 trigger, 1 room and 1 person, and log every session. Keep the trigger at half volume for days 1 to 4 and measure distance in metres rather than in guesses. The pass mark in the last column is the gate: do not move to the next row until 4 of 5 sessions hit it. Ten days is 20 sessions, enough to read the direction even if the fear is not gone, and the logic is the same one you use when you read a lung pattern on an X-ray (film reading): the comparison carries the information, not the single image.
| Days | Sessions | Exposure (sec) | Distance (m) | Pass mark |
|---|---|---|---|---|
| 1-2 | 2 | 30 | 3.0 | eats 3 treats in a row |
| 3-4 | 2 | 45 | 2.5 | pupils stay normal |
| 5-6 | 2 | 60 | 2.0 | grooms again within 60 sec |
| 7-8 | 2 | 90 | 1.0 | approaches within 30 sec |
| 9-10 | 2 | 120 | 0.5 | eats at half volume, 2 m |
When to stop the session
- The cat stops eating and refuses a treat for 30 seconds.
- Pupils stay dilated for more than 60 seconds after the trigger stops.
- Breathing climbs over 40 breaths per minute at rest.
- The cat hides for more than 10 minutes after the session.
- Any hiss, swat or growl in the first 2 minutes: end it, note the distance, restart there next time.
Five steps to run this week
- Write the trigger in 1 sentence and rate today's response from 1 to 10.
- Set session 1 at 2 out of 10, 3 metres away, 3 minutes long.
- Log recovery seconds, treat count and distance for 3 sessions before changing anything.
- Move 1 variable per session, distance or volume, never both.
- If the fear started after age 8, add a urine strip and blood work at the same visit: reading a urinalysis (urine baseline) rules out the metabolic causes before you spend 6 weeks on training.
Frequently asked
How many tries does it take before a cat stops reacting?
For a mild trigger such as a doorbell or a parked carrier, most cats settle somewhere between 6 and 10 sessions held at 2 out of 10 intensity, one session a day. The response usually halves every 2 sessions once the cat stays under threshold. If 4 sessions produce no change at all, stop increasing repetitions and look for pain, hearing loss or a second trigger in the same room.
Can one bad experience really undo weeks of training?
Yes, and it usually takes less than 10 seconds. A single startling event lowers the reaction threshold, so the next exposure produces a bigger response with less input. That is why the recovery gap matters: 48 to 72 hours of zero exposure, then restart 2 stages below the point of the incident. Pushing on the next day is the most common way owners convert a small fear into a lasting one.
Should I comfort the cat when it is scared?
Comfort does not reward fear, since fear is not an operant behaviour being reinforced, but it also does not teach the cat anything. What changes the response is exposure under threshold with a predictable end. A calm voice and a treat are useful because they mark the session as safe and keep the cat eating, which is the clearest sign that intensity is still at 2 out of 10.