Hans-Werner Hunziker


Why do we need audio memory for reading?

According to Professor B.Fischer of Freiburg University (Germany; see link below)about 50% 
of children with reading and spelling difficulties (Dyslexia) are slow in visual perception. 
Maybe this is not surprising.  
But his findings that about 45% of dislexic children have difficulties with 
audio perception are less obvious.
According to Baddeley´s concept of working memory it is clear that difficulties with the
auditive short term memory may handicap readers.


Research with the CD-ROM training program EAGLE-EYE has shown that a person uses
quite different strategies to increase scores on the 30 visual games.
In accordance with Baddeley´s concept of working memory some people use visual encoding
("I'm just looking for the group of signs that look the same")  while others prefer
auditive (phonological) encodings. These can often be observed directly when the encoding 
is done with semi-silent pronunciation.

The following example shows how visual and audio parts of working memory are used when a series
of letters has to be matched with multiple choice items:

This example holds good for someone who already knows how single letters are pronounced.
The ability to recognize whole words is ruled out in this game.


There are basically six phases that can be observed. And if the final outcome is correct, 
all of them must have worked correctly. If the final outcome is wrong, an error may have 
ocurred in one or more of these 6 phases.

The first 3 phases may be called "reading", the second 3 phases refer to "search and compare".

"Reading" is visual, then audio (phonological) and then strategic: 

1. Correct recognition of each letter (= visual)
2. Correct association with the proper sound. (= audio)
3. Using good strategies for keeping the sounds in the audio memory (phonological loop)
(which fades out after 1.5 seconds)
A strategy may be to concentrate on the intonation, to build an artificial word or to concentrate
on the crucial elements of the string.

"search and compare" is first audio, then strategic and finally visual.


1. Correct reproduction from the audio memory. Especially no sequencing errors.
2. Forming a visual representation.
3. Searching for the multiple choice answer corresponding to it.


This very complex series of skills is required for learning to read.

If there are no difficulties in all the 6 subskills we can safely assume that reading and 
spelling should work without problems.

People with dyslexia have difficulties in one or several of these subskills.
On the other hand it is possible to compensate difficulties in perception by using
creative strategies.
This could help to explain why there are famous scientists with dyslexia.

Training visual and audio perception does not mean a loss in creative strategies.
Quite the opposite is true: Playing perception games increases the motivation to develop
new perception strategies.
The reason why so-called "highly gifted" children have much better results on these games
is just their use of innovative strategies.

If a child has difficulties with audio memory a logical - and observed - strategy is
to use visual encoding. The disadvantage of this strategy: The underdeveloped audio memory is
not trained at all.


The CD-ROM training program SUPER-OWL has been developed for diagnosis and training of
audio skills which are very important for learning to read but which are normally not trained
at school in a systematic way.

Skills trained with SUPER-OWL are: Pitch, timbre and rythm sequence discrimination for musical,
natural, technical, synthetic and speech sounds.

In order to rule out word recognition, speech sounds are taken from relatively unknown
languages like Chinese, Arabic, Korean and Finnish. Missing and interchanged syllables
have to be detected by the student.

The games have been developed in such a way, that it is impossible to use visual strategies 
successfully. This ensures an actual training of the "phonological loop".



The CD-ROM SUPER-OWL contains the following games:

1.  Vibraphone 3-4***
2.  Bells 3-4
3.  Cymbals 3-4
4.  Bouzouki 3-4 
5.  Drums 4-5
6.  Vibraphone 4-5
7.  Birds 4-5
8.  Guitar 4-6
9.  Sails 4-5 synth.
10.  Rolling objects 4-5
11.  Xylophone 4 2er-3er
12.  Bells 4  3er-4er
13.  Drums 4  4er-5er  
14.  Guitar 4  4er-5er
15.  Wind-instruments 4-6
16.  Arabic 3-4 6er
17.  Finnish 4-5 6er
18.  Korean 4-6 14er
19.  Drum licks 4-6 mix
20.  Chinese 5-6 8er
21.  Chinese names 6-7 
22.  Car starter 4-6 
23.  Water 5-6
24.  Birds 6-8
25.  Waves 5-7 synth.
26.  Signals 4-6 synth.
27.  Vowels 4-6 3er
28.  Horses (stereo) 4-6
29.  Vowels 6-8 8er
30.  Drums 7-9


*** Abbreviations: 
3-4  		3 different sounds with difficulty set to "normal", 
		4 different sounds with difficulty set to "difficult"
4 4er-5er	4 different sequences with 4 elements (sounds, syllables), difficulty set to "normal", 
		4 different sequences with 5 elements (sounds, syllables), difficulty set to "difficult"
4-5 6er         4 different sequences with 6 elements, difficulty set to "normal",
		5 different sequences with 6 elements, difficulty set to "difficult"
stereo		left to right or right to left transition of noise 

synth.		Sound modified by filters and synthesizer


*Universität Freiburg - Prof. Dr. B. Fischer, AG Hirnforschung  D-79104 Freiburg,
Blicklabor