BACKGROUND
Older adults often have increasing memory problems, and worldwide about 50 million people have dementia. This syndrome gradually affects a patient over a period of 10-20 years. Intelligent virtual agents may support people suffering from memory problems.
OBJECTIVE
To identify the state of the art of experimental studies with virtual agents on a screen capable of verbal dialogues with older adults with memory problems.
METHODS
Conduct a systematic search into selected databases PubMed, SCOPUS, Microsoft Academic, Google Scholar, Web of Science and CrossRef on Virtual Agent and Memory Problems on papers that describe such experiments. Search criteria were (“Virtual Agent” OR “Virtual Assistant” OR “Virtual Human” OR “Conversational Agent” OR “Virtual Coach” OR Chatbot) AND (Dementia OR Alzheimer OR Amnesia OR “Mild Cognitive Impairment”). Risk of bias has been evaluated using the QualSyst tool that scores 14 study quality items. Eligible studies are reported in a table including country, study design type, target sample size, controls, study aims, experiment population, intervention details, results and an image of the agent.
RESULTS
Nine studies were included. The average number of participants in the studies was 18 (SD=12). The verbal interactions were generally short. The human utterance consisted in 8 out of 9 studies out of short words or phrases that were predefined in the agent’s speech recognition algorithm. The average study quality score was .68 (SD=.08) on a scale 0-1.The number of experimental studies on talking virtual agents that support people with memory problems is still small. The details on the verbal interaction are limited, which make it difficult to assess the quality of that interaction and the possible effect of confounding parameters. Further research is needed with extended and prolonged dialogues.
CONCLUSIONS
The number of experimental studies on talking virtual agents that support people with memory problems is still small. The details on the verbal interaction are limited, which make it difficult to assess the quality of that interaction and the possible effect of confounding parameters. Further research is needed with extended and prolonged dialogues.