Abstract
Background: Since his stroke 15 years ago, Tony O'Donnell has worked with BLISS, a symbolic, graphical language. BLISS has been used with a variety of clinical populations, including people with severe aphasia. O'Donnell found his adapted version of BLISS was meaningful to him when spoken and written English was not. The present study is part of a larger project that investigated the validity and generality of O'Donnell's belief that his adapted system would be useful for people with severe aphasia.
Aims: The first aim was to investigate whether people without aphasia and without training could match the adapted version of BLISS to pictures. The second aim was to conduct collaborative research involving researchers with and without aphasia and to explore their experiences.
Methods & Procedures: The performance of 55 undergraduate students was investigated on two forced-choice matching tasks: (1) matching three pictures and three BLISS strings; and (2) selecting one of two BLISS strings to match to a picture. A semi-structured interview was conducted to obtain the researchers' perceptions of the research process.
Outcomes & Results: The participants' performance across tasks showed that individuals without aphasia can match BLISS strings without training. However, there was considerable individual variation and only some participants scored significantly above chance on both tasks. The interview revealed that although this study took longer to complete, there were many positive aspects of working collaboratively. All the researchers gained personally and professionally from their involvement. Of particular interest was the realization that the researcher with aphasia had a greater ability to plan, problem solve and develop ideas than was initially thought.
Conclusions and Implications: Individuals without aphasia are able to match symbol strings to pictures without teaching. Further research is required to establish whether people with aphasia can similarly match the adapted version of BLISS to pictures. This study provides further evidence that people with aphasia can engage productively in the research process, even when experimental methodologies are employed.
Acknowledgements
The authors are grateful to Anne Edmundson and Judith Langley for helpful comments on an earlier draft of the paper. They are particularly grateful to Ann Parker who planned and conducted the interview on the research process. Pictures are from unpublished material were drawn by Eldad Druks and made available by Jane Marshall. The authors are grateful to Eldad Druks and Jane Marshall for the use of their pictures.
Declaration of Interest: The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.
Notes
1. O'Donnell (Citation2007) outlines the general properties of the adapted BLISS and shows how the language works in detail. Like all languages, the adapted BLISS is constantly evolving. For further information, Tony O'Donnell can be contacted at: e-mail: [email protected]
2. This feature of adapted BLISS was only relevant to half the experimental materials (the symbol strings corresponding to the locative states). The symbol strings corresponding to action events did reflect canonical English word order (see appendix A).
3. The symbol ‘∖’ is not used in the experimental study but we have introduced it here to illustrate some of the differences between the adapted and standard versions of BLISS.
4. There is a potential question around the probability criteria used to arrive at the threshold score for Task 1. The structure of the task meant that participants might be able to select the string with a different symbol set leaving only the reversal items to choose from, thus leaving a 1:2 probability as in Task 2. To use a 1:2 probability ratio in the calculation, we would have to be certain that the different symbol string was so obvious that every participant would be able to identify it. However, 32.59% of participants did not identify the different symbol string in Task 1. Furthermore, if the different string could be identified as such by all participants, performance on the actions items might be expected to be better than performance for locatives on Task 1. The symbol strings for actions always have the symbol representing the action in the same position in the string, with the symbols representing individual entities on either side. In the locative strings, the position of the symbols representing spatial relations varies and the symbol strings for a reversed situation are not always reversed in the string (for examples, see appendix A). Selecting the odd one out should therefore be easier for the action strings. However, participants scored higher on the locative strings than the action strings in Task 1 (see the Results section). Therefore, the results provide additional support for the use of the 1:6 probability ratio.