Effects of the SpeechEasy

This article is a summary and review of the following:
Article: Effect of the SpeechEasy on Objective and Perceived Aspects of Stuttering: A 6-Month, Phase 1 Clinical Trial in Naturalistic Environments
From: the Journal of Speech and Hearing Research, Vol. 52, April 2009
Authors: Ryan Pollard, John B. Ellis, Don Finan, Peter Ramig, University of Colorado at Boulder
This article studies the effect of the SpeechEasy under extraclinical conditions, and compares the findings with the results of previous studies.

The SpeechEasy is a device that looks like a hearing aid and feeds back the patient’s voice into his/her ear during speech, to inhibit stuttering. Delayed auditory feedback (DAF) and frequency altered feedback (FAF) are two types of altered auditory feedback (AAF). Other devices have used similar techniques to inhibit stuttering, but the SpeechEasy is the most recent and the smallest and, unlike other incarnations, can be virtually unnoticeable when worn. Also, because the rate of feedback can be adjusted to a comfortable interval, a slowed rate of speech was not necessary.

In 2003 J. Kalinowski, a PWS and the co-creator of the device, came under fire from colleagues for promoting the SpeechEasy in the media before studies appeared in peer-reviewed journals. Claims of its remarkable reduction of stuttering (up to 90%) were based on the noticed effects of AAF in laboratory environments dating back to the 1950s, and not on observations made during studies of the effects of electronic devices such as the SpeechEasy in extraclinical envirnoments.

Past studies

There have been studies since then, mainly in controlled laboratory environments, that show reduction of stuttering under certain conditions using the SpeechEasy, but not to the extent initially claimed. Significantly improved fluency was observed depending on the speech task, including monologue speech. Little extraclinical study was done, and no individual speech or self-report data was included. Effectiveness during formulated or conversational speech remained virtually unexplored.

Present study

For this study, 11 participants were involved. Samples were collected at the baseline phase (before installation of device), treatment phase (during wearing of device) and withdrawal phase (after study period when device is no longer worn). During the treatment phase articipants wore the device for 5 hours a day and visited the lab every two weeks to give speech samples – consisting of reading, conversation and asking a question of a stranger. (Speech samples from outside the laboratory were recorded using a portable digital recorder.) At the start and end of the treatment phase, participants completed self-reports consisting of surveys about general perspectives on stuttering, the impact of stuttering on one’s life, reactions to using it, etc., and gave a summary evaluation of their speech for the week in a logbook. This was to provide a fuller picture of the overall effects of the SpeechEasy.

Quantitative Results

Some of the quantitative results of the study are as follows: during oral reading there was a 58.2% reduction in stuttering during treatment phase, and 27% reduction during withdrawal phase. For conversation task there was 14.5% less stuttering during treatment phase and subsequent 6.8% reduction. In the question task, stuttering decreased 1.9% during treatment phase and 2.3% during withdrawal. When the results are combined with those of the earlier 4-month trial by Stuart et al(2004), using the treatment phase data only, stuttering reductions are reported at “22% during wearing of device in extraclinical environments.” (See article for complete results.)

Qualitative results

Attending to the second speech signal, Using fluency techniques with the device, Increased confidence and carryover effects, effects of background noise and ability to hear oneself or others were some of the qualitative information collected from the participants through a questionnaire and interview. More than half the users cited the device not being helpful in a noisy environment, or during stressful situations. Half the subjects did find it helpful in asking a question of a stranger.

As in earlier studies, the test subjects reported the sense of confidence that wearing the device gave them, and they felt they experienced an “overall improvement of fluency”. The biggest improvement in fluency was noticed when using the telephone with the SpeechEasy.

Out of the eleven participants, four did take up the offer of a 60% price reduction and purchased the SpeechEasy. Three chose not to purchase, but reported they would have continued using it if it were free. Another three subjects also opted not to purchase, and did not wish to use it even if it were free. The last participant could not be contacted at this portion of the study and was not heard from. (The missing data for this subject was accounted for in the final results).
Like other studies, this study uses a small sample (11 subjects, one of whom did not complete).This study is beneficial in that it presents data “from the first Phase 1 clinical trial of the SpeechEasy under challenging, relatively natural conditions,” and includes self-report data.

Personal Experience

This article states that proper protocols for fitting the device include using it in the ear “opposite the one used when speaking on the telephone.” However, when this reviewer purchased the SpeechEasy, it was recommended to fit it for the ear the same as hand preference, so that the user could write with the dominant hand when speaking on the phone and taking notes. Therefore, being left-handed, it was fitted for the left ear. This fitting protocol, however, contradicts the protocol stated in this article.

Right ear preference?

Because SpeechEasy users report its most positive effects when using the phone, it is natural that fitting protocols would be geared to phone useage. But should this factor be the main determinant in ear choice? What about published accounts from numerous sources describing the right ear being prefereable for interpreting language sounds? The medical evidence for this phenomenon is well documented. References are not listed here but a Google search for “right ear preference language” will result in a plethora of information and articles supporting this, including from peer-reviewed medical journals.

It is strange that right-ear preference for language hearing is not mentioned in articles on the topic of the SpeechEasy and is not a factor when getting the device – the implication is it does not matter. If it is not a factor, why isn’t it? Why isn’t the fact at least addressed in any of the literature on the topic? This is one of the most compelling issues in the application of auditory feedback devices to the problem of stuttering, a factor that future investigation into the efficacy of the Speech-Easy and other AAF devices should probably explore, or at least address.