Identifying words that emerge into consciousness: Effects of word valence and unconscious previewing

Words with negative valence capture attention and this increase in attentional resources typically enhances perceptual processing. Recently, data using continuous flash suppression (CFS) appear to contradict this. In prior research when Chinese words were unconsciously presented in CFS and contrast was raised until the word was identified, RTs to identify words with negative valence were slower than RTs to words with neutral valence. This result might be limited to situations where a logographic writing system is used and could reflect a type of cognitive aftereffect where previewing the word causes habituation. Data (N = 60) indicate that results generalize from a logographic (Chinese) to an orthographic writing system (English). In addition, when words were previewed in CFS RTs were slowed for words with negative valence relative to words with neutral valence and this was reversed when words were shown binocularly. Implications for theories of unconscious word processing and cognitive aftereffects are discussed.     

Impaired context maintenance in mild to moderately depressed students

We test the hypothesis that people with depression experience difficulties in maintaining task-relevant context information over longer periods of time using the AX version of the continuous performance task (AX-CPT). The AX-CPT requires that participants maintain a context cue (A) in an active state in order to respond correctly to a target cue (X) presented after a short delay. A total of 40 nondepressed and mild to moderately depressed students completed versions of the task with short (1-s) or long (10-s) interstimulus intervals (ISIs). Mildly depressed participants made significantly more context-dependent (BX) errors, unlike controls who made more errors on trials where good context processing would impair performance (AY). This pattern of errors was only evident in the long ISI condition, suggesting poor maintenance of contextual information.     

Comparing performance in discrete and continuous comparison tasks

The approximate number system (ANS) theory suggests that all magnitudes, discrete (i.e., number of items) or continuous (i.e., size, density, etc.), are processed by a shared system and comply with Weber's law. The current study reexamined this notion by comparing performance in discrete (comparing numerosities of dot arrays) and continuous (comparisons of area of squares) tasks. We found that: (a) threshold of discrimination was higher for continuous than for discrete comparisons; (b) while performance in the discrete task complied with Weber's law, performance in the continuous task violated it; and (c) performance in the discrete task was influenced by continuous properties (e.g., dot density, dot cumulative area) of the dot array that were not predictive of numerosities or task relevant. Therefore, we propose that the magnitude processing system (MPS) is actually divided into separate (yet interactive) systems for discrete and continuous magnitude processing. Further subdivisions are discussed. We argue that cooperation between these systems results in a holistic comparison of magnitudes, one that takes into account continuous properties in addition to numerosities. Considering the MPS as two systems opens the door to new and important questions that shed light on both normal and impaired development of the numerical system.     

Characterization of lower-limbs inter-segment coordination during the take-off extension in ski jumping

Take-off, the most important phase in ski jumping, has been primarily studied in terms of spatio-temporal parameters; little is known about its motor control aspects. This study aims to assess the inter-segment coordination of the shank-thigh and thigh-sacrum pairs using the continuous relative phase (CRP). In total 87 jumps were recorded from 33 athletes with an inertial sensor-based system. The CRP curves indicated that the thighs lead the shanks during the first part of take-off extension and that the shanks rotated faster at the take-off extension end. The thighs and sacrum first rotated synchronously, with the sacrum then taking lead, with finally the thighs rotating faster. Five characteristic features were extracted from the CRP and their relationship with jump length was tested. Three features of the shank-thigh pair and one of the thigh-sacrum pair reported a significant association with jump length. It was observed that athletes who achieved longer jumps had their thighs leading their shanks during a longer time, with these athletes also having a more symmetric movement between thighs and sacrum. This study shows that inter-segment coordination during the take-off extension is related to performance and further studies are necessary to contrast its importance with other ski jumping aspects.     

Temporal structure coding with and without awareness

In order to interpret a constantly changing environment, visual events far apart in space and time must be integrated into a unified percept. While spatial properties of invisible signals are known to be encoded without awareness, the fate of temporal properties remains largely unknown. Here, we probed temporal integration for two distinct motion stimuli that were either visible or rendered invisible using continuous flash suppression. We found that when invisible, both the direction of apparent motion and the gender of point-light walkers were processed only when defined across short time periods (i.e., respectively 100 ms and 1000 ms). This limitation was not observed under full visibility. These similar findings at two different hierarchical levels of processing suggest that temporal integration windows shrink in the absence of perceptual awareness. We discuss this phenomenon as a key prediction of the global neuronal workspace and the information integration theories of consciousness.     

Measurement of depression: Comparison between self-reports and clinical assessments of depressed outpatients

Agreement among several depression scales was investigated as regard the relative influences of administration mode (self-rating or clinical rating) and scale content. The Beck Depression Inventory (BDI), the Self-Rating Depression Scale (SDS), the Hamilton Rating Scale for Depression (HRSD), and three corresponding scales with identical structure and content but the alternative administration mode were administered to 47 outpatients with diagnoses of DSM-III major depression disorders. Correlations between the total scores and the degrees of association between corresponding items of different scales were calculated. The results suggest that differences in content contribute more to inter-scale discrepancy than differences in administration mode. The implications for the evaluation of outpatients with major depression are discussed.     

Admitting the addressee detection faultiness of voice assistants to improve the activation performance using a continuous learning framework

The main promise of voice assistants is their ability to correctly interpret and learn from user input as well as the ability to utilize this knowledge to achieve specific goals and tasks. These systems need predetermined activation actions to start a conversation. Unfortunately, the typically used solution, wake-words, force an unnatural interaction. Furthermore, this method can also confuse when the wake-word, or a phonetically similar phrase, has been said but no interaction with the system is intended by the user. Thereby, the system not only lacks the adequacy of interpersonal interaction, it moreover suffers from an addressee detection faultiness. Although various aspects have already been investigated in this field of acoustic addressee detection research, we demonstrated that the test data used so far rely on ideal conditions: The dialog complexity between human–human and human–device interactions is essentially different while in reality, the behavior of each individual addressing either another human or a device is of large variation. Thus the problem of addressee detection is simplified too much. Our approach works with a specifically designed dataset comprising of human–human and human–computer interactions of similar dialog complexity. Our proposed addressee detection faultiness framework actively communicates the system’s uncertainty that may arise. In connection with a continuous learning framework, this enables a voice assistant system to adapt itself to the users’ individual addressee behavior. This approach achieves significantly improved classification rates of 85.77%, which gives an absolute improvement of 32.22% in comparison to similar experiments employing human annotations as ground truth.     

情绪学习促进无意识信息进入意识

采用突破连续闪烁抑制范式(b-CFS), 考察了经过情绪学习的刺激在无意识中是否存在加工优势。实验1使用不同朝向的光栅刺激, 结果发现情绪学习条件比对照条件的刺激在CFS中突破抑制时间更短。实验2考察了更复杂的朝向和颜色整合的光栅刺激, 发现情绪优势效应具有特异性, 仅局限于情绪学习的整合刺激。研究表明经过情绪学习后的刺激在无意识加工中存在优势。     

Differentiating between ADHD sub-types on CCPT measures of sustained attention and vigilance

The aim of the study was to test whether sustained attention and vigilance, often considered as the same phenomenon, dissociate on the Conners' Continuous Performance Test, and whether subjects with ADHD-Inattentive type and ADHD-Combined type differ with regard to these measures. Sixty-five healthy controls and 67 subjects with ADHD between 9 and 16 years of age participated in the study. The ADHD-I group performed below control children on Hit Reaction Time Block Change, considered to measure sustained attention and the ADHD-C group scored below controls on Hit Reaction Time Inter-Stimulus-Interval, considered to measure vigilance. Comparing the two clinical groups showed a test by group interaction, with ADHD-I subjects performing below ADHD-C subjects with regard to sustained attention and above ADHD-C subjects with regard to vigilance. Sustained attention on the CCPT correlated specifically with parent and teacher ratings of inattention, but not with ratings of hyperactivity-impulsivity, while vigilance correlated with all symptom ratings.     

Incorporating Demographic Embeddings Into Language Understanding

Meaning depends on context. This applies in obvious cases like deictics or sarcasm as well as more subtle situations like framing or persuasion. One key aspect of this is the identity of the participants in an interaction. Our interpretation of an utterance shifts based on a variety of factors, including personal history, background knowledge, and our relationship to the source. While obviously an incomplete model of individual differences, demographic factors provide a useful starting point and allow us to capture some of this variance. However, the relevance of specific demographic factors varies between situations—where age might be the key factor in one context, ideology might dominate in another. To address this challenge, we introduce a method for combining demographics and context into situated demographic embeddings—mapping representations into a continuous geometric space appropriate for the given domain, showing the resulting representations to be functional and interpretable. We further demonstrate how to make use of related external data so as to apply this approach in low‐resource situations. Finally, we show how these representations can be incorporated into improve modeling of real‐world natural language understanding tasks, improving model performance and helping with issues of data sparsity.