2.2.3 ‘Automaticity’ and ‘rate of processing’ hypotheses
A proposal that attempts to address the broader picture of dyslexic functioning is that dyslexia may be caused by problems in the automatisation of skills. The concept of automatisation refers to the gradual reduction in the need for conscious control as a new skill is learned. This leads to greater speed and efficiency and a decreased likelihood of breakdown of performance under stress, as well as the ability to perform a second task at the same time with minimal disruption to either behaviour. Nicolson and Fawcett (1990, 1994) have pointed out that even highly competent dyslexic readers show a distinct lack of fluency in written language skills: their reading and writing is more laboured, more prone to error, and more susceptible to interference from other tasks. They also suggest that ‘incomplete mastery’ characterises many other features of dyslexic performance, such as problems learning to ride a bicycle or tie shoelaces. However, a general ‘automatisation deficit’ would be most evident during complex, highly demanding, multi-sensory tasks such as learning to read and write.
One way of assessing the presence of an automaticity deficit is through the use of a rapid automatised naming task (RAN). For example, individuals may be presented with a set of 50 stimuli consisting of five rows of 10 pictures of a given type in a random order and asked to name each picture as quickly as possible (see Figure 3). People with dyslexia typically show a deficit in speed on this type of task.
These results have been interpreted as further support for a phonological deficit in reading, as the task does require some phonological processing during the retrieval of the picture names. However, more recently Wolf and Bowers (1999) have suggested that difficulties in rapid naming are a separate, additional deficit to phonological difficulties, and that such a deficit is sufficient to explain reading difficulties even if the person has good phonological awareness. They suggest that people with reading difficulties fall into one of three subtypes, depending on the underlying cause of their problem:
phonology group: shows a phonological deficit, but no real problems on the RAN task
(slow naming) rate group: shows a RAN task time deficit, but no phonological problems
double deficit group: shows signs of a deficit in both phonology and naming rate and therefore has the greatest reading difficulties.
Because the most common form of treatment recommended for reading difficulties in children focuses on improving phonological awareness, evidence for different forms of cognitive deficit in dyslexia is important. If visual processing deficits do play an important role, or if there are rate and double deficit subtypes as described above, then training in phonological awareness alone would be unlikely to address all reading difficulties.
However, it is always important to look carefully at the nature of tests used to diagnose reading difficulties. RAN tasks come in two forms: serial presentation where the person is timed from start to finish, and discrete presentation where the symbols are presented one at a time and a reaction time for each item is recorded. Only serial presentation procedures are consistently associated with reading difficulties. This may be because serial tasks usually require more sustained concentration in comparison to discrete presentation tasks (where there is no need to follow a line of text or keep one's place in the grid of symbols). In other words, the apparent difference between ‘normal’ and dyslexic readers on tasks of this kind could be due to perceptual, attentional or fatigue effects rather than differences in RAN ability. This would undermine the case for a separate ‘RAN deficit’. However, it also raises a different, interesting question: why should dyslexic people be particularly susceptible to these kinds of effects? It would certainly be difficult to explain this fully in terms of a pure ‘phonological deficit’ hypothesis.
We can see how the cognitive accounts ‘explain’ many of the behavioural symptoms of dyslexia. However, even when taken together, they cannot explain dyslexia fully, nor its variability between individuals. As Frith suggests, cognitive accounts taken in isolation are incomplete: we also need to consider biological explanations.
Box 7: Definitions
Logographic strategy: A holistic approach to identifying written words via their overall visual appearance, sometimes also referred to as ‘sight word reading’.
Orthographic strategy: The approach skilled readers use to identify written words, employing both alphabetic and logographic strategies as well as their existing knowledge of grammatical forms and similar words.
Letter-sound correspondences: The associations between individual letters, and the sounds that those letters can make in a given language.
Strephosymbolia: The term coined by Orton to describe dyslexic-like symptoms (literally – ‘twisted symbols’).
Automatisation: The process of making a skill ‘automatic’, so that performance no longer needs conscious monitoring. (If fully automatised, a task can be carried out with no interference to another task being performed simultaneously.)
Rapid automatised naming task: A task requiring rapid naming of a series of letters, numbers, colours or common objects, which should involve automatic processes owing to the familiarity of the stimuli.
Serial presentation: The presentation of test items one after the other in the form of a list (or grid) that the participant has to work through in a systematic fashion.
Discrete presentation: The presentation of test items one at a time. None of the other test stimuli are visible at the same time.