Schizophrenic's 'Automatic Pilot' Still Works, But Processing New Information Causes Problems

Van Raalten first of all investigated the role of the working memory during automation. Our working memory ensures that we remember transient pieces of information such as a telephone number. Van Raalten asked study subjects, positioned in an fMRI scanner, to perform tests in which they had to remember a series of letters. Normally your working memory is then fully active. Yet the more the tests were repeated, the lower the brain activity in the areas involved in working memory function. This activity was also not compensated for by other parts of the brain involved in (long-term) memory. By automating the letter series, the study subjects therefore released working memory capacity so that it could once again process new information.

Van Raalten established that the decrease in activity was due to another function of the working memory; the restructuring of incoming information. For example, you first of all remember a telephone number as a series of independent figures. But your working memory ensures that this series is divided into chunks. Instead of 1-1-3-2-6-7-3-4-4-5, you remember 113-267-3445. In this way you only have to remember three chunks and not ten individual numbers. This characteristic ensures that once a task has become fully automated after sufficient training, it can be performed without further involvement of the working memory. As a result of this you can do other things alongside the automated task.

Not two things at once

Patients with schizophrenia can, however, process less information than healthy people. Their brains function less efficiently. It was expected that a faltering working memory in schizophrenia patients ensured that automation did not proceed well, as a result of which they could not release working memory capacity. However, tests on schizophrenia patients revealed that after training the brain activity decreased in the same manner as was the case for healthy study subjects. Although the working memory does not function well in schizophrenic patients, the automation of tasks proceeded without problems. It might therefore be expected that schizophrenia patients could also more easily perform a second task besides the first, automated task. Yet the released capacity of the working memory could not be used for a new task.

Following this 'surprise' result Van Raalten investigated where the problem could possibly be located. During new tests she discovered that the working memory in schizophrenia patients mainly struggled with the processing of information that continually changed. Consequently, schizophrenic patients may have more of a tendency to adopt automatic strategies in circumstances that demand flexible behaviour. This inability to satisfactorily process new information can lead to stereotypical behaviours, which are an important characteristic of psychiatric disorders such as schizophrenia and autism.

Van Raalten's research reveals that the working memory is not solely responsible for automation. She proposes targeting future research at other systems that interact with the working memory in automation. Furthermore, medication might also play a role in the automation performed by patients. Research is currently taking place at University Medical Centre Utrecht into the role of medication in the automation process.

Tamar van Raalten carried out her research at University Medical Centre Utrecht. Her research was funded by the Vidi grant of Nick Ramsey. In 2002, he received a Vidi grant from NWO for his research into the brain mechanisms underlying learning.