Training generalized improvisation of tools by preschool children.
Preschoolers can learn to invent tool substitutes within a group, but each new group needs its own training.
01Research in Context
What this study did
Five preschoolers with no disabilities joined the study.
The team taught each child to improvise tools only within one class.
For example, kids learned any hard object can hammer, but only hammer-type things.
They used praise and small treats when a child picked a new item that worked like the tool.
A multiple-baseline design showed the teaching started and the skill jumped.
What they found
Four of the five children quickly began grabbing new objects to do the job.
The new tool use stayed strong weeks later.
Kids only improvised inside the trained class.
When the class changed, like from hammers to scoops, they did not transfer.
You must teach each tool group on its own.
How this fits with other research
Jenkins et al. (1973) also shaped novelty. They rewarded the very first new block shape and saw creative building explode.
Pliskoff et al. (1978) adds that the same tactic works for real-world tools, not just toys.
Ayres-Pereira et al. (2018) looks like a clash. They found uneven 2-D to 3-D transfer in preschoolers, hinting kids rarely generalize.
The gap is method. Ayres-Pereira tested across very different looks (photo to object). S et al. kept the same tool function and only swapped items that already felt like hammers.
When the new item still "belongs" to the class, preschoolers transfer. When it looks and feels different, they need more teaching.
Why it matters
If you want a child to be flexible, teach the whole tool class, not just one item.
Run short drills: present the need, let the child hunt, reinforce any safe object that works.
Do the same for spoons, cups, or writing tools.
Stop after a few successes and probe again next week to be sure the skill sticks.
Plan separate lessons for each class; do not assume transfer.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Pick one tool class the child uses daily, set out three safe objects plus one odd item that still works, and reinforce any spontaneous switch.
02At a glance
03Original abstract
The development of new, "creative" behaviors was examined in a problem-solving context. One form of problem solving, improvisation, was defined as finding a substitute to replace the specifically designated, but currently unavailable, tool ordinarily used to solve the problem. The study examined whether preschool children spontaneously displayed generalized improvisation skills, and if not, whether they could be trained to do so within different classes of tools. Generalization across different tool classes was monitored but not specifically trained. Five preschool children participated in individual sessions that first probed their skill at improvising tools, and later trained and probed generalized improvisation in one or more of three tool classes (Hammers, Containers, and Shoelaces), using a multiple-baseline design. All five children were trained with Hammers, two were trained in two classes, and two were trained in all three tool classes. Four of the five children improvised little in Baseline. During Training, all five showed increased generalized improvisation within the trained class, but none across classes. Tools fabricated by item combinations were rare in Baseline, but common in Training. Followup probes showed that the training effects were durable.
Journal of applied behavior analysis, 1978 · doi:10.1901/jaba.1978.11-363