Statistical inference of minimum rank factor analysis

For any given number of factors, Minimum Rank Factor Analysis yields optimal communalities for an observed covariance matrix in the sense that the unexplained common variance with that number of factors is minimized, subject to the constraint that both the diagonal matrix of unique variances and the observed covariance matrix minus that diagonal matrix are positive semidefinite. As a result, it becomes possible to distinguish the explained common variance from the total common variance. The percentage of explained common variance is similar in meaning to the percentage of explained observed variance in Principal Component Analysis, but typically the former is much closer to 100 than the latter. So far, no statistical theory of MRFA has been developed. The present paper is a first start. It yields closed-form expressions for the asymptotic bias of the explained common variance, or, more precisely, of the unexplained common variance, under the assumption of multivariate normality. Also, the asymptotic variance of this bias is derived, and also the asymptotic covariance matrix of the unique variances that define a MRFA solution. The presented asymptotic statistical inference is based on a recently developed perturbation theory of semidefinite programming. A numerical example is also offered to demonstrate the accuracy of the expressions.This work was supported, in part, by grant DMS-0073770 from the National Science Foundation.     

Driving Under the Influence (of Stress): Evidence of a Regional Increase in Impaired Driving and Traffic Fatalities After the September 11 Terrorist Attacks

ABSTRACT— Did the September 11 terrorist attacks elicit a subsequent increase in traffic fatalities? Gigerenzer (2004) argued that decreases in flying and increases in driving in the 3 months after the attacks led to 353 "surplus" traffic fatalities. We applied a more systematic analysis to the same data and found no evidence of a significant increase in miles driven or of a significant increase in traffic fatalities. However, we did find evidence for a regional effect of the attacks on driving behaviors. We hypothesized that geographic proximity to the attacks increased stress, which in turn decreased driving quality. Our analyses revealed that in the last 3 months of 2001, the Northeast exhibited a significant increase in traffic fatalities, as well as a significant increase in fatal accidents involving an alcohol- or drug-related citation. Increased stress related to physical proximity to the attacks may explain the increase in traffic fatalities.     

Context-specific prime-congruency effects: On the role of conscious stimulus representations for cognitive control

Recent research suggests that processing of irrelevant information can be modulated in a rapid online fashion by contextual information in the task environment depending on the usefulness of that information in different contexts. Congruency effects evoked by irrelevant stimulus attributes are smaller in contexts with high proportions of incongruent trials and larger in contexts with high proportions of congruent trials (e.g., Corballis and Gratton, 2003, Lehle and Hübner, 2008). The present study investigates these context-adaptation effects in a masked-priming paradigm. Context-specific adaptation effects transfer to stimulus identities that are equiprobale in all contexts – an observation that renders explanations in terms of event-learning processes unlikely. Yet, context-specific effects vanished when the irrelevant information remained unconscious. The results suggest that context-specific adaptation of congruency effects reflect cognitive control operations that alter the processing of irrelevant information depending on the experienced utility of that information for action control.     

Transformational consumption choices: building an understanding by integrating social identity and multi‐attribute attitude theories

• Transformational value offerings result when marketers bundle products, services, and experiences in ways that provide consumers with opportunities to alter who they are. Examples include universities, healthcare providers, travel and leisure services, and many other businesses that guide individuals to change in some desired and lasting way. Pine and Gilmore ( 1999 ) predict that in the emerging economy “…transformation offerings will emerge across almost every industry that today views itself as part of the service sector” (p. 168). By definition, transformational value offerings involve identity development. Thus, it makes sense to use identity related variables to explain transformational value offering choices. This paper tests a model merging social identity theory and multi‐attribute attitude modeling to better understand factors predicting consumers' intentions to select a transformational value offering. A survey sample of United States Army Reserve recruits was used to test the integrated model. Results show that identity‐related perceptions form the basis for evaluating the value offering's benefits and intentions to adopt the offering. The prospective consumer also considers whether his or her self‐concept can accommodate that role. The proposed blending of social identity and multiattribute attitude models provides a promising framework for understanding and investigating consumers' choices to adopt transformational value offerings.

Copyright © 2009 John Wiley & Sons, Ltd.     

Anticipatory reconfiguration elicited by fully and partially informative cues that validly predict a switch in task

Task-switching studies show no behavioral benefit of partially informative cues. However, ERP evidence of an early cue-locked positivity elicited by both fully and partially informative cues suggests that both cues trigger an anticipatory component of task set reconfiguration (Nicholson, Karayanidis, Davies, & Michie, 2006). We examined this apparent discrepancy using a cued-trials task-switching paradigm with three tasks. The ERP finding of an early cue-locked positivity was replicated for both switch-to cues, which validly predicted an upcoming switch trial and specified the new task set, and switch-away cues, which validly predicted an upcoming switch trial but not the new task set. This component was not elicited by a noninformative cue that did not specify whether the task would switch or repeat. Switch-away cues resulted in more accurate but not faster responding than did noninformative cues. Modeling of decision processes confirmed a speed-accuracy trade-off between these conditions and a preparation benefit for both switch-to and switch-away cues. These results indicate that both fully and partially informative cues elicit an early anticipatory component of task set reconfiguration that is reflected in the early cue-locked positivity. We argue that the pattern of results is most consistent with a task set inhibition account of this early anticipatory component of task set reconfiguration.     

Brain and mind operational architectonics and man-made “machine” consciousness

To build a true conscious robot requires that a robot’s “brain” be capable of supporting the phenomenal consciousness as human’s brain enjoys. Operational Architectonics framework through exploration of the temporal structure of information flow and inter-area interactions within the network of functional neuronal populations [by examining topographic sharp transition processes in the scalp electroencephalogram (EEG) on the millisecond scale] reveals and describes the EEG architecture which is analogous to the architecture of the phenomenal world. This suggests that the task of creating the “machine” consciousness would require a machine implementation that can support the kind of hierarchical architecture found in EEG.