A cortical framework for invariant object categorization and recognition

In this paper we present a new model for invariant object categorization and recognition. It is based on explicit multi-scale features: lines, edges and keypoints are extracted from responses of simple, complex and end-stopped cells in cortical area V1, and keypoints are used to construct saliency maps for Focus-of-Attention. The model is a functional but dichotomous one, because keypoints are employed to model the “where” data stream, with dynamic routing of features from V1 to higher areas to obtain translation, rotation and size invariance, whereas lines and edges are employed in the “what” stream for object categorization and recognition. Furthermore, both the “where” and “what” pathways are dynamic in that information at coarse scales is employed first, after which information at progressively finer scales is added in order to refine the processes, i.e., both the dynamic feature routing and the categorization level. The construction of group and object templates, which are thought to be available in the prefrontal cortex with “what” and “where” components in PF46d and PF46v, is also illustrated. The model was tested in the framework of an integrated and biologically plausible architecture.

Facilitating word recognition: Evidence for strategic and automatic factors

An appropriate semantic context has been demonstrated to facilitate word recognition in a variety of paradigms. The present experiment examined the consequences of varying the probability that word pairs presented in a lexical decision task would be related in meaning. Early in the 20-min session, the recent density of semantic relationships between words had little influence on the size of the contextual facilitation effect, but later the influence was marked. The results suggest that the processing facilitation provided by an appropriate semantic context consisted of both a relatively automatic component and a labile strategic component with an influence that was modulated by the recent usefulness of the information provided by semantic context.     

Unified framework for recognition, localization and mapping using wearable cameras

Monocular approaches to simultaneous localization and mapping (SLAM) have recently addressed with success the challenging problem of the fast computation of dense reconstructions from a single, moving camera. Thus, if these approaches initially relied on the detection of a reduced set of interest points to estimate the camera position and the map, they are currently able to reconstruct dense maps from a handheld camera while the camera coordinates are simultaneously computed. However, these maps of 3-dimensional points usually remain meaningless, that is, with no memorable items and without providing a way of encoding spatial relationships between objects and paths. In humans and mobile robotics, landmarks play a key role in the internalization of a spatial representation of an environment. They are memorable cues that can serve to define a region of the space or the location of other objects. In a topological representation of the space, landmarks can be identified and located according to its structural, perceptive or semantic significance and distinctiveness. But on the other hand, landmarks may be difficult to be located in a metric representation of the space. Restricted to the domain of visual landmarks, this work describes an approach where the map resulting from a point-based, monocular SLAM is annotated with the semantic information provided by a set of distinguished landmarks. Both features are obtained from the image. Hence, they can be linked by associating to each landmark all those point-based features that are superimposed to the landmark in a given image (key-frame). Visual landmarks will be obtained by means of an object-based, bottom-up attention mechanism, which will extract from the image a set of proto-objects. These proto-objects could not be always associated with natural objects, but they will typically constitute significant parts of these scene objects and can be appropriately annotated with semantic information. Moreover, they will be affine covariant regions, that is, they will be invariant to affine transformation, being detected under different viewing conditions (view-point angle, rotation, scale, etc.). Monocular SLAM will be solved using the accurate parallel tracking and mapping (PTAM) framework by Klein and Murray in Proceedings of IEEE/ACM international symposium on mixed and augmented reality, 2007.     

A rapid automatic technique for generating operant key-press behavior in rats

Experimentally naive rats were trained to key press on a fixed-ratio 10 schedule of food reinforcement by a completely automatic procedure within a single, 1-hr session. Control procedures demonstrated that the resulting behavior was an operant, under control of the schedule of reinforcement and the specified reinforcing stimulus (food). A simple, combination food-tray operandum, also described, was used as the basis for the training technique.     

Contrast from social stereotypes in automatic behavior

The activation of social stereotypes can influence behavior outside of conscious awareness. It has been argued that while priming social stereotypes leads to behavioral assimilation, priming exemplars leads to behavioral contrast. Extending this theorizing, we argue that the activation of social stereotypes can also result in automatic behavioral contrast if a comparison of the self to the stereotyped group is provoked. This hypothesis is tested with speed and intellectual performance as behavioral measures. In a first study, we show that categorizing perceived others as outgroup members leads to behavioral contrast from their stereotype. The second study shows that subliminally priming the self during the activation of a stereotype to which the self does not belong leads to automatic behavioral contrast from the stereotype. These findings reverse previously found assimilation effects of social stereotype priming. Social comparison processes are discussed as a possible mediator of the results.     

The development of automatic word recognition and reading skill

The development of automatic word recognition as a function of reading skill was investigated in three experiments using the Stroop task. Reading skill level ranged from nonreaders to readers above the sixth-grade equivalent. Interference with color naming begins to emerge early in the process of learning to read, increases, and then subsequently decreases. Strings of identical letters delayed color naming for children just beginning to learn to read. The interference from words, presumably reflecting semantic processing, began developing early but did not peak until the second- to fourth-grade reading levels. These different sequences of development of interference in the various stimulus conditions suggest that word recognition is the result of a number of component processes that develop as children acquire skill in reading.     

Caudal fin branding fish for individual recognition in behavior studies

Caudal fin branding is an inexpensive technique for marking and identifying small, fragile fish. A modified tip on a hand-held soldering pencil is used to cauterize small holes in the caudal fin membrane. The technique is simple to use, appears to cause little trauma to the fish, and lasts for at least several weeks.     

From monkey-like action recognition to human language: an evolutionary framework for neurolinguistics

The article analyzes the neural and functional grounding of language skills as well as their emergence in hominid evolution, hypothesizing stages leading from abilities known to exist in monkeys and apes and presumed to exist in our hominid ancestors right through to modern spoken and signed languages. The starting point is the observation that both premotor area F5 in monkeys and Broca's area in humans contain a "mirror system" active for both execution and observation of manual actions, and that F5 and Broca's area are homologous brain regions. This grounded the mirror system hypothesis of Rizzolatti and Arbib (1998) which offers the mirror system for grasping as a key neural "missing link" between the abilities of our nonhuman ancestors of 20 million years ago and modern human language, with manual gestures rather than a system for vocal communication providing the initial seed for this evolutionary process. The present article, however, goes "beyond the mirror" to offer hypotheses on evolutionary changes within and outside the mirror systems which may have occurred to equip Homo sapiens with a language-ready brain. Crucial to the early stages of this progression is the mirror system for grasping and its extension to permit imitation. Imitation is seen as evolving via a so-called simple system such as that found in chimpanzees (which allows imitation of complex "object-oriented" sequences but only as the result of extensive practice) to a so-called complex system found in humans (which allows rapid imitation even of complex sequences, under appropriate conditions) which supports pantomime. This is hypothesized to have provided the substrate for the development of protosign, a combinatorially open repertoire of manual gestures, which then provides the scaffolding for the emergence of protospeech (which thus owes little to nonhuman vocalizations), with protosign and protospeech then developing in an expanding spiral. It is argued that these stages involve biological evolution of both brain and body. By contrast, it is argued that the progression from protosign and protospeech to languages with full-blown syntax and compositional semantics was a historical phenomenon in the development of Homo sapiens, involving few if any further biological changes.     

Observational goals and schema activation: A theoretical framework for behavior perception

A framework for behavioral observation is described in which the processing of a stream of behavior is influenced by an individual's prior knowledge about behavior, organized into cognitive “schemas.” It is proposed that observational goals determine the selection of goal-relevant schemas which, in turn, influence the observer's processing of an actor's behavior. Subjects watched a series of videotaped sequences featuring a female actor with instructions either to form an impression of her or to remember the details of the tasks she engaged in. They were then asked to answer questions that were either consistent or inconsistent with these instructions. Subjects were also instructed to unitize the actions or events that occurred in the sequence (D. A. Newtson, Journal of Personality and Social Psychology, 1973, 28 28–38). According to the unitizing measure, subjects observing with different goals attended to different behavioral features. Further, impression-forming subjects were apparently more influenced by their implicit theories of personality than were task-learning subjects when asked to give ratings of the actor's personality, and task-learning subjects were relatively more accurate in recalling task-related details. Implications of this framework and these results for research in behavioral unitization and implicit personality theory are discussed.     

An automatic recording system of the positional behavior of mice for experimental analysis

A device for automatic recording of a mouse positional behavior within a circular corridor-like cage is described. The animal’s position is successfully monitored by capacitor detection and its time spent at each position measured; the results are punched on the paper tape as binary coded numbers, and the data are processed off-line by a general-purpose computer.     

Reaction in action: intergroup contrast in automatic behavior

Whereas previous research has shown automatic behavior conforming to outgroup stereotypes, the authors demonstrate automatic behavioral contrast away from a stereotype/trait associated with an outgroup (Study 1 and 2) and point to the importance of an "us-them" intergroup comparison in this process. In Study 1, participants colored pictures more messily when neatness was associated with an outgroup rather than the ingroup. In Study 2, using a different behavior, participants primed with busy business people reacted faster than controls (assimilation) but became slower when their student ingroup identity was activated (contrast). Subliminally priming an "us-them" intergroup comparison set undermined the accessibility of outgroup stereotypic words (Study 3), especially for those higher in prejudice (Study 4). This suggests that people automatically distance themselves from outgroup attributes when intergroup antagonism is cued or chronic. Implications for the role of self and comparison processes in automatic behavior are discussed.     

Inaccurate self-knowledge formation as a result of automatic behavior

Four studies tested a post-priming misattribution process whereby a primed goal automatically influences people's behavior, but because people are unaware of that influence, they misattribute their behavior to some other internal state. People who were primed with a goal were more likely to choose an activity that was relevant to that goal, but did not recognize that the prime had influenced their choices. Instead, people used more accessible and plausible reasons to explain their behavior. The goals were seeking romantic interaction (Studies 1 and 2), helping (Study 3) and earning money (Study 4). People made choices related to these goals but misattributed the choices to temporary preferences (Studies 1 and 3) and more permanent dispositions (Studies 2 and 4). The misattribution had downstream effects, leading to choice behavior consistent with the erroneous self-knowledge. We suggest that automatic behavior can lead to a confabulated self-knowledge with behavioral consequences.     

Unraveling mechanisms for expert object recognition: bridging brain activity and behavior

Behavioral sensitivity to object transformations and the response to novel objects (Greebles) in the fusiform face area (FFA) was measured several times during expertise training. Sensitivity to 3 transformations increased with expertise: (a) configural changes in which halves of objects were misaligned, (b) configural changes in which some of the object parts were moved, and (c) the substitution of an object part with a part from a different object. The authors found that holistic-configural effects can arise from object representations that are differentiated in terms of features or parts. Moreover, a holistic-inclusive effect was correlated with changes in the right FFA. Face recognition may not be unique in its reliance on holistic processing, measured in terms of both behavior and brain activation.     

A review of environmental enrichment as treatment for problem behavior maintained by automatic reinforcement

We reviewed studies that used environmental enrichment as treatment for problem behavior maintained by automatic reinforcement. A search of behavior analytic journals produced 71 publications with a total of 265 applications of environmental enrichment used alone or in conjunction with alternative behavior manipulations (e.g., prompting, reinforcement) and problem behavior manipulations (e.g., blocking, restraint). Environmental enrichment, as a sole intervention, was efficacious in 41% of the sample. Alternative behavior manipulations, problem behavior manipulations, and a combination of both improved the overall efficacy of environmental enrichment. We discuss factors that may influence the efficacy of environmental enrichment, current trends in research on this topic, and implications for both practitioners and researchers.     

Modelling asynchrony in automatic speech recognition using loosely coupled hidden Markov models

Hidden Markov models (HMMs) have been successful for modelling the dynamics of carefully dictated speech, but their performance degrades severely when used to model conversational speech. Since speech is produced by a system of loosely coupled articulators, stochastic models explicitly representing this parallelism may have advantages for automatic speech recognition (ASR), particularly when trying to model the phonological effects inherent in casual spontaneous speech. This paper presents a preliminary feasibility study of one such model class: loosely coupled HMMs. Exact model estimation and decoding is potentially expensive, so possible approximate algorithms are also discussed. Comparison of one particular loosely coupled model on an isolated word task suggests loosely coupled HMMs merit further investigation. An approximate algorithm giving performance which is almost always statistically indistinguishable from the exact algorithm is also identified, making more extensive research computationally feasible.     

When moral identity undermines moral behavior: An integrative framework

While research bore out a reliable relationship between moral identity and moral behavior, it has not addressed potential undermining effects of moral identity on moral behavior. This paper offers an integrative framework for conceptualizing such effects. Within this framework, moral identity is defined as a goal that allows for different forms of goal pursuit. The moral identity goal can vary along three axes: prevention versus promotion orientation, external versus internal motivation, and low versus high level of abstraction. It is argued that (a) prevention orientation fosters moral disengagement, (b) external motivation promotes moral hypocrisy, and (c) a less abstract representation of the moral identity goal allows for moral licensing. These undermining effects can be overcome by promotion-oriented, internally motivated and more abstract moral identity goals.