Batch process fault detection for multi-stage broad learning system.

In the real industrial production process, some minor faults are difficult to be detected by multivariate statistical analysis methods with mean and variance as detection indicators due to the aging equipment and catalyst deactivation. With structural characteristics, deep neural networks can better extract data features to detect such faults. However, most deep learning models contain a large number of connection parameters between layers, which causes the training time-consuming and thus makes it difficult to achieve a fast-online response. The Broad Learning System (BLS) network structure is expanded without a retraining process and thus saves a lot of training time. Considering that different stages of the batch production process have different production characteristics, we use the Affinity Propagation (AP) algorithm to separate the different stages of the production process. This paper conducts research on a multi-stage process monitoring framework that integrates AP and the BLS. Compared with other monitoring models, the monitoring results in the penicillin fermentation process have verified the superiority of the AP-BLS model.

Pupillometric decoding of high-level musical imagery.

Humans report imagining sound where no physical sound is present: we replay conversations, practice speeches, and "hear" music all within the confines of our minds. Research has identified neural substrates underlying auditory imagery; yet deciphering its explicit contents has been elusive. Here we present a novel pupillometric method for decoding what individuals hear "inside their heads". Independent of light, pupils dilate and constrict in response to noradrenergic activity. Hence, stimuli evoking unique and reliable patterns of attention and arousal even when imagined should concurrently produce identifiable patterns of pupil-size dynamics (PSDs). Participants listened to and then silently imagined music while eye-tracked. Using machine learning algorithms, we decoded the imagined songs within- and across-participants following classifier-training on PSDs collected during both imagination and perception. Echoing findings in vision, cross-domain decoding accuracy increased with imagery strength. These data suggest that light-independent PSDs are a neural signature sensitive enough to decode imagination.

Pupil dilation patterns spontaneously synchronize across individuals during shared attention.

Human social behavior relies on the coupling of minds. Here we show that patterns of pupil dilations reveal mental coupling between speakers and listeners. Speakers were videotaped and eye-tracked as they discussed positive and negative autobiographical memories. An independent group of listeners were then eye-tracked while they watched these videos. As pupillary dilations reflect the dynamics of conscious attention, we computed the morphological similarity of speaker-listener pupillary time-series data as a metric of shared attention. The emotional salience of each narrative was also assessed, dynamically, by independent raters. Collective pupillary synchrony between speakers and listeners was greatest during the emotional peaks of a narrative, and decreased as narratives became less engaging. Individual differences in speaker expressivity and listener empathy revealed greatest synchrony in high expressive-high empathic dyads. Together, these findings suggest that pupillary synchrony is an implicit corollary of shared attention that can be used to track mental coupling in real time. (PsycINFO Database Record

Semantic Structural Alignment of Neural Representational Spaces Enables Translation between English and Chinese Words.

Two sets of items can share the same underlying conceptual structure, while appearing unrelated at a surface level. Humans excel at recognizing and using alignments between such underlying structures in many domains of cognition, most notably in analogical reasoning. Here we show that structural alignment reveals how different people's neural representations of word meaning are preserved across different languages, such that patterns of brain activation can be used to translate words from one language to another. Groups of Chinese and English speakers underwent fMRI scanning while reading words in their respective native languages. Simply by aligning structures representing the two groups' neural semantic spaces, we successfully infer all seven Chinese-English word translations. Beyond language translation, conceptual structural alignment underlies many aspects of high-level cognition, and this work opens the door to deriving many such alignments directly from neural representational content.

Pupil dilation patterns reflect the contents of consciousness.

The study of human consciousness has historically depended on introspection. However, introspection is constrained by what can be remembered and verbalized. Here, we demonstrate the utility of high temporal resolution pupillometry to track the locus of conscious attention dynamically, over a single trial. While eye-tracked, participants heard several musical clips played diotically (same music in each ear) and, later, dichotically (two clips played simultaneously, one in each ear). During dichotic presentation, participants attended to only one ear. We found that the temporal pattern of pupil dilation dynamics over a single trial discriminated which piece of music was consciously attended on dichotic trials. Deconvolving these pupillary responses further revealed the real-time changes in stimulus salience motivating pupil dilation. Taken together, these results show that pupil dilation patterns during single-exposure to dynamic stimuli can be exploited to discern the contents of conscious attention.

Pupil dilation dynamics track attention to high-level information.

It has long been thought that the eyes index the inner workings of the mind. Consistent with this intuition, empirical research has demonstrated that pupils dilate as a consequence of attentional effort. Recently, Smallwood et al. (2011) demonstrated that pupil dilations not only provide an index of overall attentional effort, but are time-locked to stimulus changes during attention (but not during mind-wandering). This finding suggests that pupil dilations afford a dynamic readout of conscious information processing. However, because stimulus onsets in their study involved shifts in luminance as well as information, they could not determine whether this coupling of stimulus and pupillary dynamics reflected attention to low-level (luminance) or high-level (information) changes. Here, we replicated the methodology and findings of Smallwood et al. (2011) while controlling for luminance changes. When presented with isoluminant digit sequences, participants' pupillary dilations were synchronized with stimulus onsets when attending, but not when mind-wandering. This replicates Smallwood et al. (2011) and clarifies their finding by demonstrating that stimulus-pupil coupling reflects online cognitive processing beyond sensory gain.