Effects of Model-Based and Memory-Based Processing on Speed and Accuracy of Grammar String Generation.

Abstract

Learners are able to use 2 different types of knowledge to perform a skill. One type is a conscious mental model, and the other is based on memories of instances. The authors conducted 3 experiments that manipulated training conditions designed to affect the availability of 1 or both types of knowledge about an artificial grammar. Participants were tested for both speed and accuracy of their ability to generate letter sequences. Results indicate that model-based training leads to slow accurate responding. Memory-based training leads to fast, less accurate responding and highest achievement when perfect accuracy was not required. Evidence supports participants' preference for using the memory-based mode when exposed to both types of training. Finally, the accuracy contributed by model-based training declined over a retention interval.

Developing rich and quickly accessed knowledge of an artificial grammar

“…The results of Domangue et al (2004) can be summarized very simply: Experience-based training led to fast but relatively inaccurate production, and model-based training led to slow production but more accurate knowledge (i.e., participants generated more 100% accurate strings). Integrated training led to intermediate levels of accuracy and speed.…”

“…Domangue et al (2004) found that providing participants with a representation of the underlying structure of the grammar at training can enhance accuracy but that production is quite slow. One likely reason is that GR participants had to consult a mental representation of the grammar diagram to guide their generation of exemplars.…”

“…Importantly, this cued generation test allows for assessments of level of accuracy and speed of production. Domangue, Mathews, Sun, Roussel, and Guidry (2004) recently used this paradigm to attempt to create highly accurate and quickly accessed artificial grammar knowledge by training participants with mixtures of experience-very slowly. This suggests that acquiring the underlying rules of the grammar is an important aspect of highly accurate performance.…”

The Motivational and Metacognitive Control in CLARION

“…When given partial strings, the bottom level or the top level might be used, or the two levels might work together, depending on circumstances. Based the above setup, our simulation succeeded in capturing fairly accurately human data in this task across a set of different circumstances (Domangue et al 2004). In addition, many other tasks have been simulated involving the NACS, including alphabetic arithmetic tasks, categorical inference tasks, discovery tasks, and so on.…”

“…Moreover, they were also able to complete partial strings in accordance with their implicit knowledge. The result showed that participants acquired fairly complete implicit knowledge although their explicit knowledge was fragmentary at best (Domangue et al 2004). In simulating this task, while the ACS was responsible for controlling the overall operation, the NACS was used for representing most of the relevant knowledge.…”