Limits...
The effect of target and non-target similarity on neural classification performance: a boost from confidence.

Marathe AR, Ries AJ, Lawhern VJ, Lance BJ, Touryan J, McDowell K, Cecotti H - Front Neurosci (2015)

Bottom Line: In most real-world environments there are likely to be many shared features between targets and non-targets resulting in similar neural activity between the two classes.It is unknown how current neural-based target classification algorithms perform when qualitatively similar target and non-target images are presented.This study address this question by comparing behavioral and neural classification performance across two conditions: first, when targets were the only infrequent stimulus presented amongst frequent background distracters; and second when targets were presented together with infrequent non-targets containing similar visual features to the targets.

View Article: PubMed Central - PubMed

Affiliation: Translational Neuroscience Branch, US Army Research Laboratory, Human Research and Engineering Directorate Aberdeen Proving Grounds, MD, USA.

ABSTRACT
Brain computer interaction (BCI) technologies have proven effective in utilizing single-trial classification algorithms to detect target images in rapid serial visualization presentation tasks. While many factors contribute to the accuracy of these algorithms, a critical aspect that is often overlooked concerns the feature similarity between target and non-target images. In most real-world environments there are likely to be many shared features between targets and non-targets resulting in similar neural activity between the two classes. It is unknown how current neural-based target classification algorithms perform when qualitatively similar target and non-target images are presented. This study address this question by comparing behavioral and neural classification performance across two conditions: first, when targets were the only infrequent stimulus presented amongst frequent background distracters; and second when targets were presented together with infrequent non-targets containing similar visual features to the targets. The resulting findings show that behavior is slower and less accurate when targets are presented together with similar non-targets; moreover, single-trial classification yielded high levels of misclassification when infrequent non-targets are included. Furthermore, we present an approach to mitigate the image misclassification. We use confidence measures to assess the quality of single-trial classification, and demonstrate that a system in which low confidence trials are reclassified through a secondary process can result in improved performance.

No MeSH data available.


Overall classification performance under various conditions. Left: Target vs. background distractor (T v B) discrimination performance in TO condition. Middle: Target vs. background distractor (T v B) discrimination performance in TN condition. Right: Target vs. both background distractor and non-target (T v (B+NT)) discrimination performance in the TN condition.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4544215&req=5

Figure 4: Overall classification performance under various conditions. Left: Target vs. background distractor (T v B) discrimination performance in TO condition. Middle: Target vs. background distractor (T v B) discrimination performance in TN condition. Right: Target vs. both background distractor and non-target (T v (B+NT)) discrimination performance in the TN condition.

Mentions: Overall classification performance declines when visually-similar non-target stimuli are present in the RSVP stream (Figure 4). The TO condition represents the baseline RSVP discrimination of target vs. background distractor. The classifier was highly accurate in this condition, producing average Az > 0.97. When targets are discriminated from background distractor stimuli in the TN condition (ignoring non-target stimuli) performance is not significantly different (Wilcoxon signed rank test; p = 0.06). However, when non-target stimuli are included in the discrimination, performance is significantly worse than when they were not included (Wilcoxon signed rank test; p < 0.001).


The effect of target and non-target similarity on neural classification performance: a boost from confidence.

Marathe AR, Ries AJ, Lawhern VJ, Lance BJ, Touryan J, McDowell K, Cecotti H - Front Neurosci (2015)

Overall classification performance under various conditions. Left: Target vs. background distractor (T v B) discrimination performance in TO condition. Middle: Target vs. background distractor (T v B) discrimination performance in TN condition. Right: Target vs. both background distractor and non-target (T v (B+NT)) discrimination performance in the TN condition.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4544215&req=5

Figure 4: Overall classification performance under various conditions. Left: Target vs. background distractor (T v B) discrimination performance in TO condition. Middle: Target vs. background distractor (T v B) discrimination performance in TN condition. Right: Target vs. both background distractor and non-target (T v (B+NT)) discrimination performance in the TN condition.
Mentions: Overall classification performance declines when visually-similar non-target stimuli are present in the RSVP stream (Figure 4). The TO condition represents the baseline RSVP discrimination of target vs. background distractor. The classifier was highly accurate in this condition, producing average Az > 0.97. When targets are discriminated from background distractor stimuli in the TN condition (ignoring non-target stimuli) performance is not significantly different (Wilcoxon signed rank test; p = 0.06). However, when non-target stimuli are included in the discrimination, performance is significantly worse than when they were not included (Wilcoxon signed rank test; p < 0.001).

Bottom Line: In most real-world environments there are likely to be many shared features between targets and non-targets resulting in similar neural activity between the two classes.It is unknown how current neural-based target classification algorithms perform when qualitatively similar target and non-target images are presented.This study address this question by comparing behavioral and neural classification performance across two conditions: first, when targets were the only infrequent stimulus presented amongst frequent background distracters; and second when targets were presented together with infrequent non-targets containing similar visual features to the targets.

View Article: PubMed Central - PubMed

Affiliation: Translational Neuroscience Branch, US Army Research Laboratory, Human Research and Engineering Directorate Aberdeen Proving Grounds, MD, USA.

ABSTRACT
Brain computer interaction (BCI) technologies have proven effective in utilizing single-trial classification algorithms to detect target images in rapid serial visualization presentation tasks. While many factors contribute to the accuracy of these algorithms, a critical aspect that is often overlooked concerns the feature similarity between target and non-target images. In most real-world environments there are likely to be many shared features between targets and non-targets resulting in similar neural activity between the two classes. It is unknown how current neural-based target classification algorithms perform when qualitatively similar target and non-target images are presented. This study address this question by comparing behavioral and neural classification performance across two conditions: first, when targets were the only infrequent stimulus presented amongst frequent background distracters; and second when targets were presented together with infrequent non-targets containing similar visual features to the targets. The resulting findings show that behavior is slower and less accurate when targets are presented together with similar non-targets; moreover, single-trial classification yielded high levels of misclassification when infrequent non-targets are included. Furthermore, we present an approach to mitigate the image misclassification. We use confidence measures to assess the quality of single-trial classification, and demonstrate that a system in which low confidence trials are reclassified through a secondary process can result in improved performance.

No MeSH data available.