We all make everyday errors, such as putting a top on inside out, or pouring hot water on our cereal instead into our mug for tea, but people with brain damage can make so many such errors that it interferes with their daily living.

When it comes to studying everyday mistakes or 'action errors' in healthy participants most researchers have, up until now, relied on diary methods.  Unfortunately, such studies can be unreliable because they depend on people recalling their past behaviour without bias. Meanwhile, current tests such as the Naturalistic Action Test, are only designed for use with clinical populations – indeed, healthy participants tend to score near the maximum.

Now researchers in Philadelphia have developed a novel laboratory task – an adapted version of The Coffee Challenge – which can be used to study everyday action errors in healthy participants or the mildly impaired. Tania Giovannetti at Temple University and colleagues developed the task on the basis that errors committed by brain damaged patients may be caused by a lack of processing resources, a situation they aimed to simulate in healthy people.  

Seventeen healthy participants, with an average age of 35 years, were instructed to make a cup of coffee for two fictional characters, Joe and Martha. The two cups had to be made differently in every respect – for example Joe wanted a travel mug and artificial sweetner whereas Martha required a ceramic mug and real sugar. In all, the two drinks involved the use of 16 items which were located on surfaces laid out in a U-shape around the participants.

The researchers first looked at the effects of practice. As expected, the participants' performance betrayed all the signs of becoming progressively more dependent on a learned 'action plan'. That is, they became quicker, with fewer general errors and greater error detection – they were noticing their own mistakes. Meanwhile 'anticipation errors', the premature completion of a later part of the task sequence, became more frequent, and errors of omission less frequent.

In the second experiment, the researchers tested the same participants again but this time they introduced a secondary, concurrent task, designed to simulate the loss of cognitive resources suffered by people with brain damage. The Oral Trail Making Test requires participants to count aloud in letter-number pairs, for example: A1, B2, C3. Each time the participants attempted The Coffee Challenge, they also had to carry out the Oral Trail Making test with as few errors as possible, with the counting starting each time with an arbitrary letter-number pairing (e.g. R2, S3).

Against the researchers' expectations, the concurrent task did not cause the participants' performance to betray a lack of cognitive resources in a way that resembled the kind of everyday errors made by brain damaged patients. Yes, errors on The Coffee Challenge increased, and performance slowed, but errors of omission did not increase and error monitoring remained intact.

The researchers surmised this could be because, while the participants' cognitive resources had been depleted, their action plan – their higher level knowledge of the task – had remained intact. By contrast, the kind of everyday errors observed in some brain damaged patients may result from both a loss of cognitive resources and a deficit in the task 'action plan' or task knowledge.

"This paper provides evidence that the Coffee Challenge is a valid and reliable new method for the study of action errors in healthy or mildly impaired participants," the researchers concluded. "The Coffee Challenge includes a training phase, baseline condition, and divided-attention condition that reliably elicits errors after practice. Thus, the Coffee Challenge fills an important gap in the naturalistic action methodology and appears to be a promising new tool for future research."

Another observation to emerge from the study related to the order with which the participants used the task items and could help with patient rehabilitation. On each trial, the location of the items was varied, so that participants could either stick to using their preferred ordering or they could adapt their strategy according to the layout of the items. During both the practice and divided attention parts of the study, performance was superior among those participants who stuck to using the task items in the same order. "Our analysis of serial-order consistency strongly suggest that patients undergoing rehabilitation for everyday functioning should be encouraged to learn and perform tasks in a consistent serial order to minimise error," the researchers advised.

References:

Giovannetti, T., Schwartz, M.F. & Buxbaum, L.J. (2007). The Coffee Challenge: A new method for the study of everyday action errors. Journal of Clinical and Experimental Neurospsychology, 29, 690-705. http://dx.doi.org/10.1080/13803390600932286

Weblinks:

How many D'oh moments does the average person have (from the BPS Research Digest)