An Interactive Relevance Feedback Interface for Evidence-Based Health Care

Effort-based decisions, in which people weigh potential future rewards against effort costs required to achieve those rewards involve both cognitive and physical effort, though the mechanistic relationship between them is not yet understood. Here, we use an individual differences approach to isolate and measure the computational processes underlying effort-based decisions and test the association between cognitive and physical domains. Patch foraging is an ecologically valid reward rate maximization problem with well-developed theoretical tools. We developed the Effort Foraging Task, which embedded cognitive or physical effort into patch foraging, to quantify the cost of both cognitive and physical effort indirectly, by their effects on foraging choices. Participants chose between harvesting a depleting patch, or traveling to a new patch that was costly in time and effort. Participants' exit thresholds (reflecting the reward they expected to receive by harvesting when they chose to travel to a new patch) were sensitive to cognitive and physical effort demands, allowing us to quantify the perceived effort cost in monetary terms. The indirect sequential choice style revealed effort-seeking behavior in a minority of participants (preferring high over low effort) that has apparently been missed by many previous approaches. Individual differences in cognitive and physical effort costs were positively correlated, suggesting that these are perceived and processed in common. We used canonical correlation analysis to probe the relationship of task measures to self-reported affect and motivation, and found correlations of cognitive effort with anxiety, cognitive function, behavioral activation, and self-efficacy, but no similar correlations with physical effort.

Research suggests that cognitive fatigue has a negative impact on physical activity participation. However, the mechanisms underlying this effect are yet unclear. Using an effort-based decision-making paradigm, we examined whether individuals weigh physical effort-costs more strongly when they are cognitively or physically fatigued. Twenty university students visited the lab on three occasions. On each visit, participants underwent a manipulation that was designed to either induce cognitive fatigue (i.e., 2-back task), physical fatigue (i.e., handgrip exercise), or served as a control condition (i.e., documentary watching). After the manipulations, participants performed an effort-based decision-making task in which they decided for 125 offers whether they accepted the offer to exert the required level of physical effort to obtain rewards that varied in value. The probability to accept offers declined with increasing effort requirements whereas the general probability to accept offers was not reduced by any of the experimental conditions. As expected, the decline in accepted offers with increasing effort requirements was stronger after prolonged exertion of physical effort compared to the control condition. Unexpectedly, this effect was not found after exerting cognitive effort, and exploratory analyses revealed that the impact of physical effort exertion on physical effort-based decisions was stronger than that of cognitive effort exertion. These findings suggest that people weight future physical effort-costs more strongly after exerting physical effort, whereas we could not find any evidence for this after exerting cognitive effort. We discuss multiple explanations for this discrepancy, and outline possibilities for future research.

Supervised learning algorithms (relevance feedback (RF) algorithms) are often used in content based image retrieval (CBIR) systems to enhance interactive search and browsing of image databases. One of the issues associated with RF based CBIR systems is the lack of a large training set. Labeling of images is a time consuming activity and user's usually do not have the patience to label a large set. The challenge is to somehow leverage the much larger set of unlabeled images to improve the performance of CBIR systems. In this paper we propose a novel RF algorithm which learns from both labeled and unlabeled data. Our proposed algorithm also uses active learning so as to maximize the information gained from a given amount of user feedback.

The opportunity cost of time modulates cognitive effort

Relevance Feedback is a technique used for enhancing retrieval accuracy in multimedia database systems. In this paper two novel relevance feedback algorithms are proposed for 3D object databases, in which the relative scores of various users, which express users' subjectivity, are kept accumulatively as additional descriptors. Each object is interpreted as a charged particle, whose relative scores represent the value of the charge. Based on these charges, semantic forces are calculated between the 3D objects, which are repelled or attracted properly in the feature space. The forces are of dual nature, semantic and geometric, in the first algorithm, whereas they are purely semantic in the second one. Furthermore, a novel algorithm for annotation propagation is developed, which is based on a linear prediction scheme of the changes that must be made in the feature vector of a newly added 3D object in the database, according to alterations that has already taken place to objects in the database. The combination of low and high level features in one formula, is able to fill the semantic gap as much as possible till time being, while the proposed content-free retrieval method illustrates the fact that in the long run, a purely semantic algorithm can provide excellent retrieval results.

The performance of a Content-Based Image Retrieval System (CBIR) depends on (a) the system's adaptability to the user's information needs, which permits different types of indexing and simultaneously reduces the subjectivity of human perception for the interpretation of the image visual content and (b) the efficient organization of the extracted descriptors, which represent the rich visual information. Both issues are addressed in this paper. Descriptor organization is performed using a fuzzy classification scheme fragmented into multidimensional classes, instead of the previous works where fuzzy histograms were created in one dimension using, for example, the feature vector norm. Multidimensionality relates the descriptors with one another and thus allows a compact and meaningful visual representation by mapping the elements of the resulted feature vectors with a physical visual interpretation. Furthermore, fuzzy classification is applied for all visual content descriptors, in contrast to the previous approaches where only color information is exploited. Two kinds of content descriptors are extracted in our case; global-based and region-based. The first refers to the global image characteristics, while the second exploits the region-based properties. Regions are obtained by applying a multiresolution implementation of the Recursive Shortest Spanning Tree (RSST) algorithm, called M-RSST in this paper. The second issue is addressed by proposing a computationally efficient relevance feedback mechanism based on an optimal weight updating strategy. The scheme relies on the cross-correlation measure, instead of the Euclidean distance which is mainly used in most relevance feedback algorithms. Cross-correlation is a normalized measure, which expresses how similar the two feature vectors are and thus it indicates a metric of their content similarity. The proposed scheme can be recursively implemented in the case of multiple feedback iterations, instead of the previous approaches. Furthermore, it provides reliable results regardless of the number of selected sample and the feature vector size improving relevance feedback performance, as compared to other approaches.

Successful decision making often requires weighing a given option's costs against its associated benefits, an ability that appears perturbed in virtually every severe mental illness. Animal models of such cost/benefit decision making overwhelmingly implicate mesolimbic dopamine in our willingness to exert effort for a larger reward. Until recently, however, animal models have invariably manipulated the degree of physical effort, whereas human studies of effort have primarily relied on cognitive costs. Dopamine's relationship to cognitive effort has not been directly examined, nor has the relationship between individuals' willingness to expend mental versus physical effort. It is therefore unclear whether willingness to work hard in one domain corresponds to willingness in the other. Here we utilize a rat cognitive effort task (rCET), wherein animals can choose to allocate greater visuospatial attention for a greater reward, and a previously established physical effort-discounting task (EDT) to examine dopaminergic and noradrenergic contributions to effort. The dopamine antagonists eticlopride and SCH23390 each decreased willingness to exert physical effort on the EDT; these drugs had no effect on willingness to exert mental effort for the rCET. Preference for the high effort option correlated across the two tasks, although this effect was transient. These results suggest that dopamine is only minimally involved in cost/benefit decision making with cognitive effort costs. The constructs of mental and physical effort may therefore comprise overlapping, but distinct, circuitry, and therapeutic interventions that prove efficacious in one effort domain may not be beneficial in another.

This paper presents a subspace SVM ensemble algorithm for adaptive relevance feedback (RF) learning. Our method deals with the case that user's relevance feedback examples are usually insufficient and overlapped together in feature space, which decreases the learning effectiveness of RF classifiers. To enhance classification efficiency in such case, multiple SVMs are learned by clustering-based training set partition, each of which fits its cluster-specific sample distribution and gives labeling regressions to test samples that fall within this cluster. To adapt features to sample distribution within each cluster, AdaBoost feature selection is conducted onto pyramid Haar of H&I bands in HSI space. In AdaBoost, we evaluate the feature discriminative ability by an entropy-based uncertainty criterion, based on which an Eigen feature subspace is constructed in cluster-specific SVM training. Finally, regression results of multiple SVMs are probabilistic assembled to give the final labeling prediction for test image. We compare our cluster-based cascade SVMs (CSS) RF method in COREL 5,000 database with: 1. Single SVM; 2. Active Learning SVM [5]; 3. Bootstrap Sampling SVM [7]. The superior experimental results demonstrate the efficiency of our algorithm.

The semantic gap between low level visual features and high level semantic concepts is an obstacle to the development of image retrieval. Relevance feedback techniques narrow the semantic gap to some extent. In this paper a relevance feedback algorithm is presented based on fuzzy semantic relevance matrix (FSRM). During the retrieval process, the weights in the FSRM are adjusted according to user's feedback and the FSRM are modified by learning more time. Experimental results show the effectiveness of the algorithm in the paper.

Motivation is underpinned by cost-benefit valuations where costs—such as physical effort or outcome risk—are subjectively weighed against available rewards. However, in many environments risks pertain not to the variance of outcomes, but to variance in the possible levels of effort required to obtain rewards (effort risks). Moreover, motivation is often guided by the extent to which cognitive—not physical—effort devalues rewards (effort discounting). Yet, very little is known about the mechanisms that underpin the influence of cognitive effort risks or discounting on motivation. We used two cost-benefit decision-making tasks to probe subjective sensitivity to cognitive effort (number of shifts of spatial attention) and to effort risks. Our results show that shifts of spatial attention when monitoring rapidly presented visual stimuli are perceived as effortful and devalue rewards. Additionally, most people are risk-averse, preferring safe, known amounts of effort over risky offers. However, there was no correlation between their effort and risk sensitivity. We show for the first time that people are averse to variance in the possible amount of cognitive effort to be exerted. These results suggest that cognitive effort sensitivity and risk sensitivity are underpinned by distinct psychological and neurobiological mechanisms.

An exploration of diversified user strategies for image retrieval with relevance feedback

Relevance feedback is an iterative search technique to bridge the semantic gap between the high level user intention and low level data representation. This technique interactively determines a user's desired output or query concept by asking the user whether certain proposed 3D models are relevant or not. For a relevance feedback algorithm to be effective, it must grasp a user's query concept accurately. In this paper, we propose a relevance feedback framework based on Hidden Markov Model Regression (HMMR) in content-based 3D model retrieval systems. Given a 3D model retrieval system, we collect and store user's feedback and use HMMR to enhance the retrieval performances. Experimental results show that this algorithm achieves higher search accuracy than traditional query refinement schemes.

The reward after-effect of effort expenditure refers to the phenomenon that previous effort investment changes the subjective value of rewards when obtained. However, the neural mechanisms underlying the after-effects of effort exertion are still not fully understood. We investigated the modulation of reward after-effects by effort type (cognitive vs. physical) through the lens of neural dynamics. Thirty-two participants performed a physically or cognitively demanding task during an effort phase and then played a simple gambling game during a subsequent reward phase to earn monetary rewards while their electroencephalogram (EEG) was recorded. We found that previous effort expenditure decreased electrocortical activity during feedback evaluation. Importantly, this effort effect occurred in a domain-general manner during the early stage (as indexed by the reward positivity) but in a domain-specific manner during the later and more elaborative stage (as indexed by the P3 and delta oscillation) of reward evaluation. Additionally, effort expenditure enhanced P3 sensitivity to feedback valence regardless of effort type. Our findings suggest that cognitive and physical effort, although bearing some surface resemblance to each other, may have dissociable neural influences on the reward after-effects.

Objective: Hearing impairment is linked to increased fatigue, yet little is known about the real-world impact of this fatigue. This qualitative study investigated the experience of daily-life fatigue in people with a hearing impairment.Design: Individual face-to-face semi-structured interviews were conducted. Thematic analysis was then used to analyse the data.Study sample: Fourteen hearing impaired participants (aged 44–70 years) who varied in terms of hearing loss, hearing aid status, age and gender.Results: The themes and sub-themes that emerged from the transcripts were: Fatigue (effort-driven fatigue, emotion-driven fatigue, breaks and recovery, the perceived relationship between hearing impairment and fatigue, and sleep), Effort (cognitive effort and physical effort), Coping Strategies (withdrawal, avoidance and planning), Relationships and Emotions, Hearing Aid Impact.Conclusions: The study highlights that hearing impairment-related fatigue is experienced by many but not all, and to different extents. Hearing aids were weakly linked to a beneficial effect on fatigue. In addition to the more widely researched effort-driven fatigue, participants described fatigue linked to the negative emotions related to having a hearing impairment. These findings, in conjunction with the widespread utilisation of different coping strategies, demonstrate that the experience of fatigue is varied and likely dependent on personal factors and lifestyle.

Neuroscientific methods and tools can be used to discover unique and useful consumer insights. This chapter provides an overview of the various uses for neuroscientific methods for consumer research. It outlines key neuroscientific methods and tools, including functional magnetic resonance imaging, electroencephalography, various biometric methods, transcranial magnetic stimulation, transcranial direct current stimulation, and neurogenetic association approaches. A key consideration with neuroscientific methods is that each one provides different neurophysiological information. Like other neuroscientific methods, biometrics is often used in studies to capture unconscious responses that cannot be captured with traditional marketing research methods. The most common biometric methods are: electrodermal activity, facial coding, heart rate monitoring, and eye tracking. Heart rate monitoring uses an electrocardiograph machine to obtain real-time measures of arousal, attention, cognitive effort, or physical effort. With continued improvements in neuroscientific technologies, as well as data analytics techniques, there will be greater adoption of neuroscientific methods.