AAC COMMUNICATION RATE MEASUREMENT:
TOOLS AND METHODS FOR CLINICAL USE

Barry A. Romich* and Katya J. Hill**
* Prentke Romich Company
Wooster, OH 44691
** Edinboro University of Pennsylvania
Edinboro, PA 16444

ABSTRACT
Augmentative and alternative communication (AAC) rate enhancement is a significant goal for all working in this area of assistive technology. Automated language activity monitoring (LAM) performance tools allow the calculation of communication rate based on logged data from the use of actual AAC systems by people who rely on them. A standard method for calculating peak and average communication rate in the clinical setting is proposed. Implications are significant in the areas of clinical intervention, outcomes measurement, and research.

BACKGROUND
The goal of AAC intervention is the most effective communication possible for the individual. Generally, effective communication is accomplished through spontaneous novel utterance generation (SNUG). Measuring AAC communication rate can provide valuable information on the efficacy of various rate-enhancement strategies to improve interactive communication. However, we make decisions about the benefits of available strategies without much assistance from a research base (1). The content of this paper focuses on communication using assistive technology.

For people who rely on AAC, communication rate almost always is far slower than normal speech. Yet communication effectiveness can be a significant factor in determining personal achievement in life. Therefore, every effort must be made to maximize communication rate for this population.

The recent development of automated logfiles (2) and the AAC language activity monitor (LAM) (3) for clinical use has made available quantitative data on which to base intervention decisions. The LAM has made possible various analyses of data that can prove useful to clinicians as well as to people who rely on AAC. Analysis of AAC language samples collected using LAM has provided information similar to that available from samples of speaking individuals. This includes measures of vocabulary diversity and linguistic complexity. In addition, LAM data, because it includes language event time stamps, has allowed identification of language representation method usage.

OBJECTIVE
The development of LAM makes practical for the first time the quantitative measurement of communication rate as part of the clinical intervention process. However, there has been no standardized definition of or method of calculating communication rate for people who rely on AAC. Further, no tools have been available to automate the measurement of communication rate. The objective of this work has been to develop a standardized method of rate measurement.

METHOD
This paper proposes a method of calculating communication rate. Communication rate generally is reported in terms of words per minute. This proposal is based on those units. Language samples from LAM are reported in the following model:

20:37:00 “I need ”
20:37:05 “*[VOLUME UP]*”
20:37:06 “*[VOLUME UP]*”
20:37:07 “*[VOLUME UP]*”
20:37:14 “something ”
20:37:16 “to drink ”
20:37:19 “i”
20:37:20 “m”
20:37:24 “m”
20:37:28 “ediately ”

The time stamp is a 24 hour format with one second resolution. Items included inside the *[ - ]* are non-language events. The first process applied to raw LAM data is often the editing of the data into utterances, e.g. “I need something to drink immediately ”. This prepares the data for analyses using commercially available or custom computer-based language transcript analysis tools.
The proposal presented here is that communication rate measurement be based on SNUG utterances and that the language sample include at least ten such utterances. While the use of preprogrammed messages by people who rely on AAC is not uncommon, the inclusion of these items in a rate measurement calculation would be misleading. This is based in part on analysis of LAM pilot study logged data that shows use of preprogrammed utterances to be under 1% of total AAC system communication from high-end users on technology that supports all language representation methods (4).

Both peak and average rates may be of interest. Peak is being defined as the highest rate for a particular utterance in which the number of words following the first event exceeds the corresponding mean length of utterance (MLU) for the entire sample. Average rate is calculated on the basis of all utterances included in the language sample and is weighted according to the number of words in the utterance. Here is a listing of steps to be taken to convert raw LAM data into peak and average words per minute.

1. Edit LAM sample into utterances.
2. Remove non-language data and adjust the time stamps accordingly.
3. Remove preprogrammed utterances.
4. Calculate words per minute for each utterance.
A. Start time (S) = time of first event in the utterance (one second resolution).
B. End time (E) = time of last event in the utterance (one second resolution).
C. Words (W) = words after first event in the utterance.
D. Words per minute (WPM) = (W / (E – S)) X 60.
5. Words per minute peak (WPMP) = Highest WPM of an utterance longer than the sample MLU.
6. Words per minute average (WPMA) = Average WPM, weighted by number of words.

For the example above (I want something to drink immediately ), the first event (I need ) occurred on the minute (00). This is the start time (S). Seven seconds were used to adjust the volume. Thus “something “ occurred an adjusted seven seconds (14 – 7) into the utterance. The utterance concludes with “ediately “ at an adjusted 21 seconds (28 – 7) into the utterance. So the end time (E) is 21 seconds. The utterance contains six words, but two of them were the first event. This leaves the word count (W) at four. Rate for this utterance is then calculated:
WPM = (4 / (21 – 0)) X 60 = 11.4 words per minute.

This utterance has four words following the first event. If the sample MLU (following the first event) is under 4.0, this utterance would be included in the identification of peak rate. For the average rate calculation each utterance is weighted according to the number of words (following the first event) in the utterance. This results in a more representative measure of overall communication performance than the true average.

DISCUSSION
The availability of standardized methods for measuring and calculating communication rate has implications in the areas of clinical intervention, outcomes measurement, and research. The end result of the use of these tools is the enhanced communication and higher personal achievement of people who rely on AAC.

By having measurement tools and methods, additional information on the acquisition of personal AAC performance will be forthcoming. Of great value will be an improved understanding of the distinction between systems that are easy to use at first encounter and those that provide the most effective communication in the long run. Understanding the characteristics of the learning process should impact AAC assessment approaches.

Through the use of standardized methods based on quantitative logged data, the field of AAC is making the transition from an art to a science.

REFERENCES
1. Beukelman DR & Miranda P (1998). Augmentative and Alternative Communication: Management of Severe Communication Disorders in Children and Adults. 2nd Edition. Baltimore: Paul H. Brookes Publishing Co.

2. Higginbotham DJ & Lesher GW (1999). Development of a voluntary standard format for augmentative communication device logfiles. Proceedings of the RESNA ’99 Annual Conference. Long Beach, California. pp 25-27.

3. Romich, BA & Hill, KJ (1999). A language activity monitor for AAC and writing systems: Clinical intervention, outcomes measurement, and research. Proceedings of the RESNA ’99 Annual Conference. Long Beach, California. pp 19-21.

4. Hill K & Romich B (1999). AAC Automated Language Activity Monitoring. American Speech-Language-Hearing Association Convention (ASHA). San Francisco, California.

ACKNOWLEDGEMENT
The National Institute for Deafness and Other Communication Disorders of NIH has provided funding to Prentke Romich Company to support the work on LAM.

Barry A. Romich, P.E.
Prentke Romich Company
1022 Heyl Road
Wooster, OH 44691
Tel: 330-262-1984 ext. 211
Fax: 330-263-4829
Email: bromich@aol.com