Table 1.
Behavioural data outcomes.
Figure 1.
ERP waveforms evoked by target-present hits (Attended Hits) and by target-absent correct-rejections (Unattended Correct-rejections).
1A) (i) ERP waveforms evoked by Attended Hits in low- and high-distraction conditions. The N1 was maximal at Cz in low-distraction at 118 ms and in high-distraction at 120 ms. It was attenuated in high-distraction (t14 = 2.649; P = 0.019). 1A (ii) Isopotential maps of Attended Hits N1 peak difference between low- and high-distraction at 120 ms. 1B) (i) ERP waveforms evoked by target-absent correct rejections (Unattended Correct-rejections) in low- and high-distraction conditions. The N1 was maximal at Cz in low-distraction at 120 ms and in high-distraction at 122 ms. It was attenuated in high-distraction (t14 = 2.387; P = 0.032). 1B (ii) Isopotential map of Unattended Correct-rejections N1 peak difference between low- and high-distraction at 120 ms. 1C Reclassified ERP waveforms evoked by target absent correct-rejections (i.e. Unattended Correct-rejections) in high-energy/high-distraction, low-energy/high-distraction and low-distraction at electrode Cz. No difference was found between high-energy and low-energy high-distraction trials (t14 = 0.022; P = 0.983). Comparisons between low-energy/high-distraction and low-distraction revealed a significant difference (t14 = 2.336; P = 0.035). Thus, N1 attenuation in high-distraction is not due to energetic masking associated with the speech distractor.
Figure 2.
Inter-trial phase coherence and Phase Distributions.
2A) Time-frequency plots of grand-averaged Inter-trial phase coherence at electrode Cz for Attended Hits in low (i) and high (ii) distraction. (iii) Time-frequency plot and (iv) FDR thresholded map of the differences between distraction conditions (low minus high) in Inter-trial phase coherence 2B) Time-frequency plots of grand-averaged inter-trial phase coherence at electrode Cz for Unattended Correct-Rejections in low (i) and high (ii) distraction. (iii) Time-frequency plot and (iv) FDR thresholded map of the differences between distraction conditions (low minus high) in inter-trial phase coherence. There was a decrease of theta/alpha inter-trial phase coherence around the N1 latency in high-distraction for both Attended Hits and Unattended Correct-rejections. There was a decrease of theta and alpha inter-trial phase coherence for Unattended Correct-rejections (but not Attended Hits) at approximately 300 to 400 ms post-stimulus in high-distraction. 2C) Grand-averaged radial histograms of phase angle distributions in the 150 ms/6 Hz time-frequency bin in low- and high-distraction for Attended Hits. Mean phase angles for low- and high-distraction are indicated by the blue and red lines, respectively. The distribution of phase angles was rotated (delayed) by distraction. The difference in mean phase angles was marginally significant (F(1,14) = 3.06; P = 0.09) and the difference in phase concentration was significant (c2 (1, n = 15) = 4.56; P = 0.03). 2D) Grand-averaged radial histograms of phase angle distributions for the 150 ms/6 Hz time-frequency bin in low- and high-distraction for Unattended Correct-rejections. The difference in mean phase angles and phase concentrations were both significant (F(1,14) = 12.35; P = 0.0015) and (c2 (1, n = 15) = 5.11; P = 0.02), respectively, for Unattended Correct-rejections. Note that high-distraction in both Attended Hits and Unattended Correct-rejections appears to both broaden and shift the distribution of phases of 6 Hz theta band signals.
Figure 3.
Simulated ERP Waveforms and Phase Distributions.
3A) Attenuate and Delay model. A single-cycle 6 Hz (theta band) sinusoidal waveform embedded in 1/f noise (omitted for clarity) was simulated for three levels of modulation: 100% amplitude/0 ms fixed delay; 80% amplitude/20 ms fixed delay; 60% amplitude/40 ms fixed delay. In this model the waveform on individual trials within each condition varied in amplitude but had fixed latencies. 3B) Distraction Decoherence Model. A single-cycle 6 Hz (theta band) sinusoidal waveform embedded in 1/f noise (omitted for clarity) was simulated for three levels of jitter: 100% amplitude/no jitter; 100% amplitude/20 ms mean jitter; 100% amplitude/40 ms mean jitter. In this model, the waveform on individual trials was always 100% amplitude for each condition but varied in latency. 3C) Radial phase distributions and mean phase at the N1 latency for the Attenuate and Delay model (i) 100% amplitude/0 ms delay; (ii) 80% amplitude/20 ms fixed delay; (iii) 60% amplitude/40 ms fixed delay. Mean phase angles are indicated by the red lines. 3D) Radial phase distributions and mean phase at the N1 latency for the Distraction Decoherence Model. (i) 100% amplitude/0 ms delay; (ii) 100% amplitude/20 ms mean jitter; (iii) 100% amplitude/40 ms mean jitter. Note that in both models, the distribution of phases is broadened and rotated counter-clockwise (i.e. delayed).
Figure 4.
Time-frequency analysis of the Attenuate and Delay and Distraction Decoherence models.
4A) Attenuate and Delay Model: time-frequency plots of (i) inter-trial phase coherence (ii) total power (iii) evoked power and (iv) induced power for the 100% amplitude/0 ms delay, 80% amplitude/20 ms fixed delay, and 60% amplitude/40 ms fixed delay modulations, respectively. (v, vi) Wilcoxen Rank Sum test masked for time-frequency bins that showed a significant directional cross-over interaction between evoked power and induced power: (v) compares 100% amplitude/0 ms delay to 80% amplitude/20 ms fixed delay and (vi) compares 80% amplitude/20 ms fixed delay to 60% amplitude/40 ms fixed delay. Light blue indicates time/frequency bins with p-values between 0.05 and 0.01 and green indicates bins with p-values less than 0.01. 4B) Distraction Decoherence Model: time-frequency plots of (i) inter-trial phase coherence (ii) total power (iii) evoked power and (iv) induced power for the 100% amplitude/0 ms delay, 100% amplitude/20 ms mean jitter and 100% amplitude/40 ms mean jitter modulations. (v, vi) Wilcoxen Rank Sum test masked for time-frequency bins that showed a significant directional cross-over interaction between evoked power and induced power: (v) compares 100% amplitude/0 ms delay to 100% amplitude/20 ms mean jitter and (vi) compares 100% amplitude/20 ms mean jitter to 100% amplitude/40 ms mean jitter. Note that the test for the cross-over interaction between evoked power and induced power selectively identifies the phase jitter built into the Distraction Decoherence model without falsely finding phase jitter in the Attenuate-and-Delay model.
Figure 5.
Directional Cross-Over Interactions Differentiate Attenuate and Delay from Distraction Decoherence Models.
5A) Attenuate and Delay Model: Grand-averaged evoked (i) and induced (ii) power (125 ms to 225 ms and from 4 Hz to 8 Hz) at 100% amplitude/0 ms delay and 80% amplitude/20 ms fixed delay modulations. Grand-averaged evoked (iii) and induced (iv) power (125 ms to 225 ms and from 4 Hz to 8 Hz) at 80% amplitude/20 ms fixed delay and 60% amplitude/40 ms fixed delay modulations; error bars indicate the standard error of the mean. 5B) Distraction Decoherence Model: Grand averaged evoked (i) and induced (ii) power (125 ms to 225 ms and from 4 Hz to 8 Hz) at 100% amplitude/0 ms delay and 100% amplitude/20 ms mean jitter modulations. Grand averaged evoked (iii) and induced (iv) power (125 ms to 225 ms and from 4 Hz to 8 Hz) at 100% amplitude/20 ms mean jitter and 100% amplitude/40 ms mean jitter modulations. Note the presence of an evoked power by induced power directional cross-over interaction in the Distraction Decoherence Model but not in the Attenuate and Delay Model.
Figure 6.
Decoherence Due to Distraction.
6A) Time frequency plots of (i) total power (ii) evoked power and (iii) induced power for Attended Hits in low (above) and high (below) distraction. (iv) Wilcoxen Rank Sum maps masked to show bins exhibiting a significant directional cross-over interaction between evoked and induced Power. Light blue indicates time/frequency bins with p-values between 0.05 and 0.01 and green indicates bins with p-values less than 0.01. 6B) Time frequency plots of (i) total power (ii) evoked power and (iii) induced power for Unattended Correct-rejections in low (above) and high (below) distraction. (iv) Wilcoxen Rank Sum maps masked to show bins exhibiting a significant directional cross-over interaction between evoked power and induced power. Note the significant crossover interaction in the theta/alpha band at the N1 latency range, particularly for Unattended Correct-rejections.
Figure 7.
Evoked Power by Induced Power Directional Cross-over Interaction due to Distraction.
7A) Grand-averaged evoked (i) and induced (ii) power in low- and high-distraction for Attended Hits at time-frequency bins: 125 to 150 ms; 6 to 8 Hz; error bars indicate the standard error of the mean. 7B) Grand-averaged evoked (i) and induced (ii) power in low- and high-distraction for Unattended Correct-rejections at time-frequency bins: 125 to 150 ms; 6 to 8 Hz. Note that both Attended Hits and Unattended Correct-rejections show evidence of a directional evoked power by induced power cross-over interaction.