Response-initiated imaging of operant behavior using a digital camera.
A thirty-dollar webcam can photograph each response and reveal when reinforcement is making movement stereotyped.
01Research in Context
What this study did
Iversen (2002) wired a thirty-dollar webcam to a lever box. Each press snapped a photo.
The camera caught the rat’s exact pose at the moment of response.
Photos were stored by the same program that counted the lever presses.
What they found
When every press paid off, the rats settled into one stiff, repeated pose.
When payment stopped, the poses quickly became varied again.
The snapshots showed the change in form faster than the cumulative record did.
How this fits with other research
PLISKOFF (1963) did the same trick with an oscilloscope-camera forty years earlier. That setup filmed lights that moved with each response. Iversen (2002) swaps the scope for a cheap webcam and stores the shots on a PC.
Cook et al. (2019) later showed that removing reinforcement again and again inside one session brings back old, varied topographies. Iversen (2002) caught the first proof in still photos.
Lejuez et al. (2001) used a touch screen to vary pay rate and saw harder-to-disrupt responding at richer rates. Iversen (2002) adds the visual: rich pay locked in one stiff movement, not just more responses.
Why it matters
You can plug a webcam into your next single-case setup and get free topography data. One look at the photo series tells you if the movement is becoming rigid or staying flexible. That visual check can warn you a behavior is about to become stereotyped before your graph shows a change.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Tape a webcam to the side of your work table, set it to snap on switch press, and scroll the photos after the session to see if the child’s hand path looks the same each time.
02At a glance
03Original abstract
A miniature digital camera, QuickCam Pro 3000, intended for use with video e-mail, was modified so that snapshots were triggered by operant behavior emitted in a standard experimental chamber. With only minor modification, the manual shutter button on the camera was replaced with a simple switch closure via an I/O interface controlled by a PC computer. When the operant behavior activated the I/O switch, the camera took a snapshot of the subject's behavior at that moment. To illustrate the use of the camera, a simple experiment was designed to examine stereotypy and variability in topography of operant behavior under continuous reinforcement and extinction in 6 rats using food pellets as reinforcement. When a rat operated an omnidirectional pole suspended from the ceiling, it also took a picture of the topography of its own behavior at that moment. In a single session after shaping of pole movement (if necessary), blocks of continuous reinforcement, in which each response was reinforced, alternated with blocks of extinction (no reinforcement), with each block ending when 20 responses had occurred. The software supplied with the camera automatically stored each image and named image files successively within a session. The software that controlled the experiment also stored quantitative data regarding the operant behavior such as consecutive order, temporal location within the session, and response duration. This paper describes how the two data types--image information and numerical performance characteristics-can be combined for visual analysis. The experiment illustrates in images how response topography changes during shaping of pole movement, how response topography quickly becomes highly stereotyped during continuous reinforcement, and how response variability increases during extinction. The method of storing digital response-initiated snapshots should be useful for a variety of experimental situations that are intended to examine behavior change and topography.
Journal of the experimental analysis of behavior, 2002 · doi:10.1901/jeab.2002.77-283