Abelson, R. P., Psychological status of the script concept, American Psychologist, 1981, 36, 715-729

---

Concept of schema dates back to Piaget.

People in artificial intelligence find scripts useful. In this context, scripts are series of expected events.

Sometimes low-level inferences are used instead of scripts. A script govens a body of inferences. A set of expectations like ordering ina a restaurant can also be seen as a vignette. The author found that priming one script improved recognition of another script. Strong scripts have more rigid ordering than weak scripts.

"In sum, a script is a hypothesized cognitive structure that when activiated organizes comprehension of event-based situations".

Scripts seem to help "fill the gaps" in comprehension. Some studies have shown that in recall of described events with part of the script left out, people falsely recall the missing script events. People seem to rememer the generic script with some specific modifications. Events reordered in a common script are recalled as being closer in order to the general script.

Scripts as Performative Structures
There are scripts in understanding and scripts in performance. However, to begin a script a person must have a stable cognitive representation of the script, there must be an evoking context, and the person must commit to begin the script. There is an "action rule" that tells the person when to begin the script.

Issues like bystander intervention can be understood in terms of action rules and criteria. The foot in the door technique could also be seen as a learned action rule facilitatitng scripted behavior the second time.

A helping script might be activiated by a request that requires legitimacy. The authors study of a woman asking someone to mail a package so she could catcha train ( 80% compliance) and so she could go shipping ( 20% compliance). A "would you do me a favor" request was 55%45% compliance under the same request reasons.

Action and behavior are often contradictory. LaPiere sucessfully travelled with two Chinese companions in hotels across the US. But when he called them afterward and asked if it was OK for Chinese to stay at their hotel, 90% said no.

Attitudes learned when in direct contact with an object or issue correlate better to behavior than attitudes learned without direct involvement. Another study showed that thinking about an issue before filling out an attitude scale results in a better corelation with behavior than without prior thought. Even putting a mirror in front of people increases correlation.

We have seen the convergence of animal-conditioning behavior with cognitive information processing.

Eight factors that embody script fundtions:
* equifinal actions (many actions reach same goal)
* variables give predictive generality (what is ordered is what is served)
* paths (branch points offering alternatives to normal procedures -- menu onthe table)
* scene selection
* tracks (different script variants)
* interferences (obstacles and errors)
* distraction (from playing script)
* free behaviors (conversation in a restaurant)

One could regard a script as a psychlogical category, and events during the script as attributes. Metascripts are superordinates of scripts. Many of its scenes are defined at a higher level of abstraction than those of a script. Schrank proposes MOP (memory organization packets) that serve to organize scenes and scripts for later use. The activities of the present serve to trigger recall of scripted behavior in the past that can help aid comprehension and behavior.

Some scripts have internal or external coherence between them. Scripts with a high level of coherence lend themselves to being combined into a metascript. Tomkin has referred to a nuclear scene, one that people experience repeatedly, almost inevitably.