Table 1
By-participant mean response times to correct responses on the vibrotactile task and mean accuracies at that task, as well as mean word recognition accuracies in the easy and hard SNRs and A-only and AV modalities for Experiment 1. Standard deviations are indicated in parentheses.
| SNR | Accuracy on vibrotactile task (%) | Response time to vibrotactile task (ms) | Word recognition accuracy (%) |
|---|---|---|---|
| Easy | 79.88 (10.89) | 1,045 (180) | 91.04 (3.46) |
| Hard | 74.54 (12.82) | 1,222 (226) | 46.69 (6.12) |
Figure 1
By-participant response times to the secondary vibrotactile task for the easy (left) and hard (right) conditions. The dot represents the mean response time in each condition, and the shape of each plot represents the distribution of responses across participants.
Table 2
By-participant mean response times to correct responses on the vibrotactile task and mean accuracies at that task, as well as mean word recognition accuracies in the easy and hard SNRs and A-only and AV modalities for Experiment 2. Standard deviations are indicated in parentheses.
| SNR | Accuracy on vibrotactile task (%) | Response time to vibrotactile task (ms) | Word recognition accuracy (%) | |||
|---|---|---|---|---|---|---|
| AV | A-only | AV | A-only | AV | A-only | |
| Easy | 79.80 (10.03) | 77.90 (8.86) | 1,047 (197) | 998 (161) | 97.21 (2.13) | 92.22 (3.20) |
| Hard | 75.35 (11.43) | 72.34 (10.83) | 1,111 (195) | 1,089 (203) | 75.15 (5.97) | 39.96 (8.21) |
Table 3
Coefficients, standard errors, t- or z-values, and p-values for the accepted models. t-values are reported for the continuously valued response times and z-values are presented for the binomially distributed word recognition accuracy data. See available code for details of the random effects structure for each model.
| Model | Fixed effect of interest | β | SE | t or z | p |
|---|---|---|---|---|---|
| Response time ~ SNR + modality + lipreading | SNR | –76.91 | 14.32 | –5.37 | <.001 |
| modality | 33.64 | 12.82 | 2.62 | .01 | |
| Response time ~ SNR + modality + lipreading + SNR * modality | SNR * modality | 28.76 | 6.71 | 4.29 | <.001 |
| Planned comparisons, Easy SNR | |||||
| modality | 48.09 | 19.43 | 2.48 | .02 | |
| Planned comparisons, Hard SNR | |||||
| modality | 21.25 | 19.34 | 1.10 | .28 | |
| Response time ~ SNR + modality + lipreading + SNR * modality * lipreading | SNR * modality * lipreading | 295.70 | 50.81 | 5.82 | <.001 |
| Planned comparisons, Easy SNR | |||||
| modality * lipreading | 55.23 | 149.02 | 0.37 | .71 | |
| Planned comparisons, Hard SNR | |||||
| modality * lipreading | –265.27 | 143.80 | –1.85 | .07 | |
| Word recognition accuracy ~ SNR + modality + SNR * modality | SNR | 3.88 | 0.09 | 41.16 | <.001 |
| modality | 1.94 | 0.08 | 24.18 | <.001 | |
| SNR * modality | –1.06 | 0.11 | –9.59 | <.001 | |
| Vibrotactile accuracy ~ SNR + modality | SNR | 0.29 | 0.05 | 5.49 | <.001 |
| modality | 0.17 | 0.05 | 3.47 | <.001 | |
Figure 2
By-participant response times to the secondary vibrotactile task (A) and word recognition accuracy (B) in the easy and hard SNRs in the A-only and AV modalities in Experiment 2. The dot represents the mean response time (A) or accuracy (B) in each condition, and the shape of each plot represents the distribution of responses across participants.
Figure 3
Model predictions and 95% confidence intervals for average response times to the vibrotactile task as a function of lipreading score, SNR, and modality. Each point represents the average response time for a given condition and participant.
Figure 4
By-participant recall of the 3- and 4-back words (A) and overall word recognition accuracy (B) in the easy and hard conditions in the A-only and AV modalities in Experiment 3. The dot represents the mean recall (A) or accuracy (B) in each condition, and the shape of each plot represents the distribution of responses across participants.
Table 4
By-participant mean recall and speech recognition accuracy in the easy and hard SNRs as well as the A-only and AV modalities for Experiment 3. Standard deviations are indicated in parentheses.
| SNR | Recall accuracy (%) | Word recognition accuracy (%) | ||
|---|---|---|---|---|
| AV | A-only | AV | A-only | |
| Easy | 62.03 (15.75) | 62.00 (12.51) | 96.87 (2.07) | 92.01 (2.63) |
| Hard | 52.80 (13.55) | 52.88 (13.67) | 73.34 (6.77) | 43.41 (6.83) |
[i] Note: Recall accuracy refers to the 3- and 4-back words, but word recognition accuracy refers to the average accuracy for all words.