Infant Method

The procedure

Paul Smeets and his colleagues tested this procedure for infant potty training on four children, aged 3-6 months (Smeets et al 1985). Infants were trained by their own parents. To accommodate busy parents, researchers accepted that infants would skip training on some days. A “training day” was defined as a day when parents spent at least 2 hours training their infants.

Before training began, parents observed their infants to learn what body signals precede voiding. Then they progressed through three phases of training.

Phase I.

The goal of Phase I was to teach the baby to associate his own body signals with using the potty. Parents kept babies within visual range of a potty chair throughout the session. When parents judged that babies were ready to eliminate, they tapped the potty chair to get their babies attention, then sat their babies on the potty. If babies eliminated within 3 minutes of being placed on their chairs, parents showed their enthusiastic approval. Otherwise, parents removed their babies from the chairs and tried again later (when the babies showed more signals). When babies had accidents, they were changed without a show of emotion. Parents continued these sessions until their babies had at least 18 bowel movements in the potty chair and had experienced at least 8 out of 10 training days without bowel accidents (Smeets et al 1985).

Phase II.

Babies were kept within 30 cm of potty chairs (by placing baby and potty chair in a playpen, for instance). When the babies showed body signals of elimination, babies were prompted to touch or grab the potty chair. If babies didn’t touch the chair after they were verbally prompted, parents gently guided their hands. Then babies were placed on the potty chair as in Phase I (described above). Babies were also placed on the potty chair if they spontaneously touched or grabbed the chair, or if they showed no signs but were overdue for a voiding. Phase II continued until babies started to reach for the potty without prompting for more than 50% of the trials for 4-6 days running. Babies were also required to have a low rate of accidents—no more than 10% of the total eliminations.

Phase III.

Training continued as in Phase II, except that the potty chair was moved farther away (up to 4 meters, depending on the baby’s crawling skills) and parents were told to “keep the number of daily prompts to a minimum” (Smeets et al 1985). Training ended when babies had completed 15 out of 18 consecutive training days without an accident and at least 80% of reaching/grabbing responses were followed by eliminations.

As noted above, all four infants completed this infant potty training program before they were 12 months old–before any of them could walk. Moreover, no negative side effects were reported.


Smeets PM, Lancioni GE, Ball, TS, and Oliva DS. 1985. Shaping self-initiated toileting in infants. Journal of applied behavior analysis, 18: 303-308.