Methodology :
Two inpatients(Patient A:M/44; Patient B:F/56) having HIV-associated progressive multifocal leukoencephalopathy encountered difficulties in practicing/improving their app-based N-back performance. This was potentially due to app constraints on stimulus presentation time, required reaction time, alongside reduced attention/working memory being the key feature of patients’ cognitive condition (Zucchella et al., 2011).
A tailor-made multi-channel compensatory training strategy was adopted with training conducted manually using screen-capped stimulus, and speed and immediate feedback adjusted according to patient’s condition. To enhance patients’ encoding and memory rehearsal, the N-back task was broken-down to:(1)naming the stimuli; (2)verbalizing the answers; (3)giving behavioral responses.
On-site N-back training was conducted while patients received inpatient rehabilitation services. Outcomes included accuracy(%) and reaction time(seconds) for the app-based training and subjective feedback.
Result & Outcome :
Patient A, having severe cognitive impairment, exhibited preservative errors and showed no improvement in app-based 1N-back performance despite attending>10 weekly on-site training sessions. There was significant immediate improvement in accuracy when comparing before (M=66.71%, SD=4.70), and after (M=97.40%, SD=3.16) adopting the compensatory training strategy for various stimuli, t(38)=-23.67, p< .001. There was no speed-accuracy trade-off as there was no significant change in reaction time when comparing before (M=1.07 sec, SD=0.29) and after the strategy (M=0.99 sec, SD=0.13). The highest level of 7 was achieved in 1N-back for various stimuli after 19 weekly on-site training sessions.
Patient B, characterized by mild-to-moderate cognitive impairments, failed to respond within the required time-frame, leading to low accuracy (M=76.19%, SD=4.27), with reaction time (M=1.89 sec, SD=0.17), during app-based 2N-back despite attending>10 weekly on-site training sessions. After adopting the compensatory strategy, there was significant immediate improvement in accuracy for various stimuli (M=90.72%, SD=4.35), t(17)=-6.51, p< 0.001, with reduced mean reaction time (M=1.68 sec, SD=0.08), t(17)=2.50, p< 0.05. The strategy was rated 10/10 for usefulness, and was subsequently applied to recalling phone numbers in voicemail, leading to reduced frequency of re-listening.
This study highlights the possibility of utilizing tailor-made compensatory strategies to assist patients with significant neuropsychological impairments when practicing app-based N-back training. Patients also learnt to apply these strategies in daily life. The limitations of the case study and lack of objective measurements for generalizing training performance in daily life are noted.
