Model Selection for Unsupervised Learning of Visual Context

This study addresses the problem of choosing the most suitable probabilistic model selection criterion for unsupervised learning of visual context of a dynamic scene using mixture models. A rectified Bayesian Information Criterion (BICr) and a Completed Likelihood Akaike’s Information Criterion (CL-AIC) are formulated to estimate the optimal model order (complexity) for a given visual scene. Both criteria are designed to overcome poor model selection by existing popular criteria when the data sample size varies from small to large and the true mixture distribution kernel functions differ from the assumed ones. Extensive experiments on learning visual context for dynamic scene modelling are carried out to demonstrate the effectiveness of BICr and CL-AIC, compared to that of existing popular model selection criteria including BIC, AIC and Integrated Completed Likelihood (ICL). Our study suggests that for learning visual context using a mixture model, BICr is the most appropriate criterion given sparse data, while CL-AIC should be chosen given moderate or large data sample sizes.     

NMDA Receptor Modulation of Incidental Learning in Pavlovian Context Conditioning.

Rats exposed to a footshock show conditional fear when reexposed to the shock context. Immediate presentation of shock after placement in the context significantly reduces this fear. Preexposure to the context in the absence of shock, coupled with a minimum preshock interval during training, overcomes this immediate shock deficit. Because rats learn about the context during preexposure and express that learning after being reinforced, the context preexposure effect is an aversive analogue of latent learning. The authors examined the effect of the N-methyl-D-aspartate (NMDA) receptor antagonist D,L-2-amino-5-phosphovalerate (APV) on the facilitatory effect of context preexposure. Rats were preexposed to a chamber after APV administration. The next day they were placed in the same chamber without drug and received shock 35 s later. APV blocked the facilitatory effect of preexposure. Therefore NMDA receptors are important for contextual latent learning. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A Relational Model of Workplace Victimization: Social Roles and Patterns of Victimization in Dyadic Relationships.

This article proposes a model of the ways in which dyadic interactions between employees who occupy 1 of 4 archetypal social roles in organizations can lead to either episodic or institutionalized patterns of victimization. The model shows how the occurrence of victimization involving these 4 role types is influenced by organizational variables such as power differences, culture, and access to social capital. The model integrates behavioral and social structural antecedents of victimization to develop a relational perspective on the dynamics of harmful behavior in the workplace. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Implicit Leadership Theories in Applied Settings: Factor Structure, Generalizability, and Stability Over Time.

The present empirical investigation had a 3-fold purpose: (a) to cross-validate L. R. Offermann, J. K. Kennedy, and P. W. Wirtz's (1994) scale of Implicit Leadership Theories (ILTs) in several organizational settings and to further provide a shorter scale of ILTs in organizations; (b) to assess the generalizability of ILTs across different employee groups, and (c) to evaluate ILTs' change over time. Two independent samples were used for the scale validation (N1 = 500 and N2 = 439). A 6-factor structure (Sensitivity, Intelligence, Dedication, Dynamism, Tyranny, and Masculinity) was found to most accurately represent ELTs in organizational settings. Regarding the generalizability of ILTs, although the 6-factor structure was consistent across different employee groups, there was only partial support for total factorial invariance. Finally, evaluation of gamma, beta, and alpha change provided support for ILTs' stability over time. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Douglas W. Bray (1918-2006).

Presents an obituary for Douglas W. Bray, the inventor of the modern-day assessment center, a method used by thousands of organizations around the world to identify the best people for critical roles and to guide individuals in optimizing their talents. Bray's death in Englewood, New Jersey, on May 9, 2006, ended the extraordinary career of a pioneer in industrial/organizational (I/O) psychology who optimized the interplay of research and application. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Generation and context memory.

Generation enhances memory for occurrence but may not enhance other aspects of memory. The present study further delineates the negative generation effect in context memory reported in N. W. Mulligan (2004). First, the negative generation effect occurred for perceptual attributes of the target item (its color and font) but not for extratarget aspects of context (location and background color). Second, nonvisual generation tasks with either semantic or nonsemantic generation rules (antonym and rhyme generation, respectively) produced the same pattern of results. In contrast, a visual (or data-driven) generation task (letter transposition) did not disrupt context memory for color. Third, generating nonwords produced no effect on item memory but persisted in producing a negative effect on context memory for target attributes, implying that (a) the negative generation effect in context memory is not mediated by semantic encoding, and (b) the negative effect on context memory can be dissociated from the positive effect on item memory. The results are interpreted in terms of the processing account of generation. The original, perceptual-conceptual version of this account is too narrow, but a modified processing account, based on a more generic visual versus nonvisual processing distinction, accommodates the results. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Family-Responsive Interventions, Perceived Organizational and Supervisor Support, Work-Family Conflict, and Psychological Strain.

This study examined organizational family-responsive policies, perceptions of the organization as family supportive, and supervisor support as issues that may be salient to the experience of conflict between paid employment (work) and family roles. Data were collected from 355 managerial personnel in New Zealand. Although work-family conflict and psychological strain were strongly linked, the availability of organizational policies had no significant association with levels of conflict or strain, whereas policy usage was related only to work-to-family interference and not to family-to-work interference. On the other hand, perceptions of the organization as family supportive and supervisor support for work-family balance displayed significant relationships with key variables, highlighting the importance of these variables for interventions designed to ameliorate the negative impact of work-family conflict on managerial well-being. Implications for the effective implementation of family-responsive interventions are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The Amygdala Modulates Hippocampus-Dependent Context Memory Formation and Stores Cue-Shock Associations.

Preexposing rats to the context facilitates subsequent contextual fear conditioning. This effect depends on the hippocampus (J. W. Rudy, R. M. Barrientos, & R. C. O'Reilly, 2002). The authors report that inactivating the basolateral region of the amygdala (BLA) by injecting muscimol, a GABAA agonist, before or after preexposure reduced this effect. In contrast, bilateral injections of anisomycin, a protein synthesis inhibitor, into BLA did not impair the consolidation of the context memory. However, when injected after fear conditioning, anisomycin impaired consolidation of both contextual and auditory-cue fear conditioning. Results are consistent with 2 ideas about the amygdala's contribution to memory: (a) It modulates memory formation in other regions of the brain, and (b) it is a storage site for cue-shock associations. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Is the Promissory Note of Personality as Vulnerability to Depression in Default? Reply to Zuroff, Mongrain, and Santor (2004).

In contrast to D. C. Zuroff, M. Mongrain, and D. A. Santor (2004), the current authors find the promissory note of dependency-sociotropy (DEP-SOC) and self-criticism-autonomy (SC-AUT) as a model of risk for depression to be in default. The authors propose reorganizing what has been cast as unitary effort into 3 distinct endeavors: a psychoanalytic clinical theory, development of a refined empirical model of risk for clinical depression, and research examining the effects of DEP-SOC and SC-AUT on interpersonal relationships in nonclinical samples. The authors identify some issues that need to be accommodated regardless of whether the assessment of Zuroff et al. (2004) or their own is accepted. DEP-SOC and SC-AUT are best construed as correlated, continuous dimensions. Future work also needs to accommodate depression as chronic recurrent condition, advances in developmental psychopathology, and more stringent criteria for positing a risk factor for clinical disorder. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Student acceptance of unique meals

Students (N = 167) from the 4th, 8th, and 11th grades sorted meals, listed on individual cards, according to whether they would or would not eat them. Students were told to either consider eating the meals at home or at school. Forty-two different meals were selected from a pool of unique meals created by students in a previous study. The meals differed in head component, the presence or absence of vegetables, and the number of different foodstuffs. On average, approximately half of the meals were deemed acceptable, but older students accepted more meals than younger ones, boys accepted more meals than girls, and meals to be consumed at home were accepted more than those to be consumed at school. In addition, students tended to accept meals comprised of a small number of foodstuffs that did not contain vegetables. The also chose more meat-based meals over those containing fish or soup. These results suggest that students have rather unique food preferences that do not correspond well with nutritional standards. In order to address this problem we recommended that students be given more good choices in the school cafeteria through the introduction of a two-meal system and salad bars.