Language Learning Impairment: Integrating Research and Remediation by Paula Tallal

Language Learning Impairment:

Integrating Research and Remediation

by Paula Tallal, Ph.D.

Two independent research studies report that language learning impaired (LLI) children improved by approximately two years after only four weeks of intensive exposure to speech and language listening exercises presented with an acoustically modified speech signal, together with a new form of adaptive computer training (Merzenich, 1996; Tallal, et al., 1996).
These papers report the development of novel remediation techniques that grew out of a collaboration between Drs. Paula Tallal and Michael Merzenich. Research by Tallal and colleagues has demonstrated that LLI children require hundreds of milliseconds (msec) between acoustic events to discriminate between them, while children of the same age and intelligence level only need tens of msec. We have further shown that:

This basic sensory integration deficit interferes with the ability of LLI children to discriminate the critical brief acoustic cues within syllables and words that distinguish one phoneme from another, and

That temporal integration thresholds are highly correlated with the degree of receptive language impairment in younger LLI children, and later with these children's difficulty learning phonological decoding skills for reading.

The work of Dr. Merzenich is based on neurophysiological studies with monkeys. Dr. Merzenich and colleagues' landmark studies shows that even in adult animals, the neurons in the brain that map sensory events are highly "plastic." That is, the brain can reshape its neural circuitry in response to environmental experiences. Specifically, they showed that through intensive training, monkeys could gradually improve their identification of faster and faster sounds. When they analyzed the brains of these monkeys, they found that specific auditory regions had reorganized and significantly expanded their neural circuits.
Merzenich and Tallal hypothesized that the training exercises used for neural plasticity studies with monkeys might be adapted to alleviate some of the sensory processing constraints that have been reported in LLI children. In our earlier research, we had also observed that extending the duration of the brief, rapidly changing transitional elements within the acoustic wave form of speech syllables resulted in significantly improved speech discrimination of those syllables for LLI children. We hypothesized that if we could create a computer algorithm to acoustically disambiguate the rapidly changing acoustic cues within ongoing fluent speech, this might help LLI children process, and thus consistently represent, phonological cues in syllable, word and sentence context. Once represented correctly at a high level of accuracy with modified speech, the goal would be to drive processing through successive adaptive training, closer and closer to normal levels. To test these hypotheses, we developed two basic methodologies:

Computer "games" designed to adaptively speed up the rates of processing of acoustic cues within both non-speech and speech stimuli, and

Listening exercises designed to explicitly train on-line phonological discrimination and language comprehension using acoustically modified speech.

Computer games were developed using both nonverbal and verbal stimuli. The games were adaptive. By adaptive, we mean that the stimulus sets and series of trials were controlled by each subject's trial by trial performance. The adaptive computer games were developed with the aim of first establishing the precise acoustic parameters within stimulus sets required for each subject to maintain 80% correct performance on that stimulus set. Once that threshold point was determined for each subject, the subject's own performance determined the acoustic parameters of each subsequent trial. The goal of the training was to first determine the thresholds for specific acoustic variables and then, for subjects with elevated thresholds, attempt to drive them to process closer and closer to a more normal processing rate. The "games" were designed to be fun for the subject and to maintain ongoing attention.
The papers in Science report the results of the first two studies we completed using these new techniques. In each study, subjects were trained with the acoustically modified speech and computer "games" three hours per day, five days a week for six weeks. The results of Study One were dramatic. They demonstrated that improvement in processing rate, coupled with training with acoustically modified speech, resulted in significant improvement in speech and language. From a clinical perspective, these exciting results clearly hold promise for a new and highly effective intervention for children with central auditory processing phonological analysis and/or language comprehensive problems.
Study Two replicated the results of Study One, and in addition included a highly controlled additional treatment group, who received the same treatment, but with normal rather than acoustically modified speech and without adaptive training aimed at speeding up the rate of acoustic processing. As predicted, improvement made by the modified speech training group was significantly greater than that made by the control group that received essentially the same training, but with natural, unmodified speech.
How do we think this new therapeutic program works? First, the adaptive computer "game" training was shown to be highly effective in significantly speeding up the rate of acoustic processing. Indeed, as a group, thresholds went from 400 msec pre-training to 100 msec post-training. For some LLI children, thresholds post-training approached the normal tens of millisecond range. We anticipate that with longer training times and more trials, these new training exercises may continue to drive these children's processing rates into the normal range. The data from our first two studies demonstrate that faster processing rates and improved speech discrimination abilities translate into significantly improved ability to process the individual sounds within words - a fundamental goal of both speech and language therapy for language impaired children, as well as phonological awareness training for reading impaired (dyslexic) children. These studies show that LLI children clearly are able to better distinguish one speech sound from another. After training, these children appear to have been able to set up distinct, (not "fuzzy") phonological representations for each phoneme in their language. Further, the results of our follow-up studies demonstrate that the improvements in phonological processing and language comprehension are enduring. That means that the child is now able to approach speech, language and reading with a neural processing system more attuned to adequate on-line phonological processing, that is, more like that of children who are learning normally.
From a clinical perspective, data to date, from over 4,000 children completing this new training program, show that dramatic, replicable, ongoing improvements in central auditory processing, speech and language skills can be achieved in relatively short, but intensive program using these new computerized training techniques.

References:

Paula Tallal, Steve L. Miller, Gail Bedi, Gary Byma, Xiaoqin Wang, Srikantan S. Nagarajan, Christoph Schreiner, William M. Jenkins, and Michael M. Merzenich. "Language Comprehension in Language-Learning Impaired Children Improved with Acoustically Modified Speech." Science, 5 January, 1996, Vol.271, pp. 81-84.

LLI children received extensive daily training over a four week period, with listening exercises in which all speech was translated into a synthetic speech processing algorithm. They also received extensive daily training with computer "games" designed to adaptively drive improvements in temporal processing thresholds. Significant improvements in speech discrimination and language comprehension abilities were demonstrated in two independent groups of LLI children.

P. Tallal, G. Saunders, S. Miller, W.M. Jenkins, A. Protopapas and M.M. Merzenich. "Rapid training-driven improvement in language ability in autistic and other PDD children." Society for Neuroscience, Vol. 23, p. 490

A report of major success in applying adaptive training procedures disguised as computer games to 5- to 12-year-old specifically language impaired children. Seven hierarchical exercises were designed to improve aural phonetic reception in these children, and to generalize their improved aural reception skills to all aspects of language. With training, speech reception was markedly clarified and language comprehension thereby improved.

In extension, training was applied to a population of 28 pervasively developmentally disabled (PDD) children (10 autistic; 18 NOS). Children worked at the same seven computer-guided adaptive training exercises for 100 minutes per day for 20-60 days. Most PDD children made major gains in acoustic and phonological reception and in language comprehension, as measured by highly significant progressions in training exercise performance. Mean Z-score improvements in standard pre- vs post-training of these abilities (e.g., Token Test, GFW) were about 1.75. In parallel, CELF and TOLD language battery quotients improved by >1 SD in about 80% of trained PDD children. Interestingly, large improvements in both receptive and expressive battery quotients were recorded; Z-score changes averaged 1.3 for receptive LQ's and 1.1 for expressive LQ's, respectively.
These studies show that major gains in language abilities can be very rapidly achieved in at least most of these severely impaired children by computer-guided training targeting fundamental acoustic and speech reception abilities. Research supported by Scientific Learning Corporation.

S. Miller, M.M. Merzenich, G. Saunders, W.M. Jenkins and P. Tallal. "Improvements in language abilities with training of children with both attentional and language impairments." Society for Neuroscience, Vol. 23, p.490.

A large proportion of language impaired (LI) children have attentional deficit disorders (ADD) and hyperactivity (ADHD). This study examined whether there are differences in the abilities of ADD-LIs, ADHD-LIs and LIs to achieve improved aural speech reception and been successfully applied to LIs (Science 271:78-84,1996).
The study was conducted with 106 children who were identified as ADD and LI, compared with nearly 400 LI children that had no ADD. Major and equal gains in speech and language reception and usage were recorded in ADD-LI and control LLI children, documented by progressions in performance recorded in 7 adaptive, computer-based training exercises ("Fast ForWord" program), and by standard prevs post-training tests of speech and language reception, comprehension, and usage. Zscores improved by a mean of 1.6 with training for ADD-LI children on the GFW Test of Auditory Discrimination; and by a mean of 1.6 on the Token Test. TOLD and CELF language battery quotients improved in parallel: positive Z-score changes for receptive quotients were > 1.2 and 0.9 for the TOLD and CELF; expressive language quotients had average Z-score gains >0.9 and 1.0). Improvements on all CELF and TOLD standard scores and quotients were significant for both ADD-LIs and LIs at p > 0.001.
Training-induced gains in speech and language for LI vs ADD-LI children did not differ. Compliance at 100-minute-long daily training exercises over a 20-60 day long training period was equivalent. Gains of ADHDs did not differ to ADDs without H, by all measures.

Paula Tallal, Ph.D. and Michael Merzenich, Ph.D., et al. "Fast ForWord™ Training for Children with Language-learning Problems: Results from a national field study by 35 independent facilities." Paper presented at the 1997 annual meeting of the American Speech-Language-Hearing Association, Boston, MA., November 21, 1997.

Paula Tallal, Steve Miller, Bill Jenkins and Mike Merzenich. "The Role of Temporal Processing in Developmental Language-Based Learning Disorders: Research and Clinical Implications." To appear in Foundations of Reading Acquisition, Benita Blachman, Ed. Lawrence Erlbaum Assoc., Inc.

About the Author

Dr. Tallal is Professor and Co-Director of the Center for Molecular and Behavioral Neuroscience at Rutgers University, Newark, New Jersey. Dr. Tallal can be reached at 197 University Ave., Newark, NJ 07102. (201) 648-1080. Via e-mail at tallal@axon.rutgers.edu.

Copyright © July 1998 New Horizons for Learning, all rights reserved.
http://www.newhorizons.org
E-mail: info@newhorizons.org
For permission to redistribute, please go to:
New Horizons for Learning Copyright and Permission Information