No description available.

No description available.

No description available.

This quasi-experimental field study examined the efficacy of an intervention to decrease alcohol abuse by college students. The harm reduction approach states that the ultimate goal when dealing with an unsafe behavior should be abstinence, but any change in behavior in the direction of less harm is supported. This approach was used as the basis of the current research, in combination with differential reinforcement in order to reduce alcohol consumption and its behavioral outcome, blood alcohol concentration (BAC). A total of 409 male and female college students participated while in the applied setting of four fraternity parties. The study took place at two separate fraternity houses, a control fraternity and an experimental fraternity. During the intervention phase, participants with a BAC below .05 were entered into a raffle to win a cash prize. Upon entry to the intervention party, participants were given flyers announcing the raffle and contingency, and gender-specific nomograms to aid in BAC self-monitoring. Dependent measures were blood alcohol concentration measured by hand-held breathalyzers, percentage of participants below criterion BAC levels (i.e., .05 and .08), accuracy of BAC self-estimation, number of negative outcomes due to excessive alcohol consumption, number of positive outcomes due to abstinence or moderate alcohol consumption, and amount of reported fun experienced at the party. Results showed the intervention did not significantly reduce the intoxication of participants or increase the percentage of participants below criterion BAC levels. These results are best explained by a floor effect, as the experimental fraternity had a relatively low baseline BAC. The use of nomograms at the intervention party increased the accuracy of students' BAC self-estimations. Implications for nomogram use and improvements for future implementation of the incentive/reward intervention are discussed. / Master of Science

The effects of non-differential reinforcement and differential reinforcement on problem behaviors and accuracy of responding of autistic children was examined. In experiment 1, one child with autism participated, and in experiment 2, two children with autism participated. In the non-differential reinforcement condition both prompted and unprompted responses were reinforced. In the differential reinforcement condition only unprompted responses were reinforced. Overall, problem behaviors were more frequent in the non-differential reinforcement condition. In experiment 1, accuracy was higher in the differential reinforcement condition, while experiment 2 showed inconclusive results with regards to accuracy. It is concluded that non-differential reinforcement can decrease problem behaviors in teaching situations, but may not be sufficient to ensure acquisition of target tasks.

Two groups of human subjects pressed buttons on five different variable-interval (VI) reinforcement schedules presented for seven minutes each for 15 sessions. At session 16, the same VI schedules were programmed concurrently in each session either with or without a 5 s changeover delay (COD). The same schedule-correlated stimuli were employed in single and concurrent conditions. Two other groups responded on concurrent VI VI conditions from the first session with or without the COD. Response allocations under concurrent scheduling better approximated relative reinforcement frequencies when the COD was programmed. Subjects with single VI histories failed to match response and time allocations to reinforcement ratios better than subjects given no such history. Bidirectional cumulative records are discussed as a molecular data analysis technique.

Negative reinforcement can be a powerful tool for behavior analysts, yet it is often overlooked as a treatment method. Pryor (1999) outlines a method for approaching a "timid" animal using a combination of negative reinforcement and positive reinforcement. When the animal stands still, the human operates a clicker, and then retreats from the animal. Gradually, the human moves closer to the animal through the clicking and retreating shaping process. Once the human is standing close enough, food may be offered as a positive reinforcer, and the negative reinforcer is canceled out. The purpose of this study was to experimentally demonstrate the click-retreat technique with cows. A multiple-baseline design across subjects was used to test this technique. Results show that the click and retreat technique was effective. Results are discussed in terms of the difference between the click-retreat technique and systematic desensitization.

Pigeons were trained on a zero-delay identity match-to-sample task. Experiment 1 started with every correct match reinforced with grain access and subsequent conditions include higher fixed-ratio values. Experiment 2 included the same fixed-ratio values as experiment 1 with and without a limited hold (LH) on the opportunity to select a comparison stimulus. Prior research suggested that trials after reinforcement would have an increased likelihood of error, and that these errors would be reduced in LH conditions. Results confirmed this expected error pattern and in most LH conditions errors were reduced early in the ratio.

Positive and negative reinforcement contingencies have been compared in terms of preference, but the differential effects of positive and negative reinforcement on reaction time and accuracy with other variables controlled remain unclear. Fifteen undergraduate students participated in a sound discrimination task that involved random mixed-trial presentations of positive and negative reinforcement contingencies. The participants' goal was to correctly identify whether the tone was shorter or longer than 600 milliseconds. On positive reinforcement trials, the participants received feedback and money tallies only if they identified the sound length correctly, with each correct response in the positive reinforcement trials earning the participant 10 cents. On negative reinforcement trials, the participants received feedback and money tallies only if they identified the sound length incorrectly, with incorrect trials subtracting 10 cents from the participants' total money (which began at $4.00 to equalize the weights of the positive and negative reinforcement contingencies). Accuracy analyses showed a relatively curvilinear relationship between the number of errors for each participant and the binned duration of the sound stimulus, with no differences across the positive and negative reinforcement conditions. Results also indicated weak linear negative correlations at the single subject level between comparison stimulus duration and reaction time, with similar slopes between positive and negative reinforcement trials, and strong curvilinear correlations at the group level, indicating differences between grouped and individual analyses. Overall our results appear to support abandoning the distinction between positive and negative reinforcement as two separate behavioral processes.

<p>This thesis targets the orchestration challenge of the Wearable Internet of Things (IoT) systems, for optimal configurations of the system in terms of energy efficiency, computing, and  data transmission activities. We have firstly investigated the reinforcement learning on the  simulated IoT environments to demonstrate its effectiveness, and afterwards studied the algorithm  on the real-world wearable motion data to show the practical promise. More specifically, firstly,  challenge arises in the complex massive-device orchestration, meaning that it is essential to  configure and manage the massive devices and the gateway/server. The complexity on the massive  wearable IoT devices, lies in the diverse energy budget, computing efficiency, etc. On the phone  or server side, it lies in how global diversity can be analyzed and how the system configuration  can be optimized. We therefore propose a new reinforcement learning architecture, called boosted  deep deterministic policy gradient, with enhanced actor-critic co-learning and multi-view state?transformation. The proposed actor-critic co-learning allows for enhanced dynamics abstraction  through the shared neural network component. Evaluated on a simulated massive-device task, the proposed deep reinforcement learning framework has achieved much more efficient system  configurations with enhanced computing capabilities and improved energy efficiency. Secondly, we have leveraged the real-world motion data to demonstrate the potential of leveraging  reinforcement learning to optimally configure the motion sensors. We used paradigms in  sequential data estimation to obtain estimated data for some sensors, allowing energy savings since  these sensors no longer need to be activated to collect data for estimation intervals. We then  introduced the Deep Deterministic Policy Gradient algorithm to learn to control the estimation  timing. This study will provide a real-world demonstration of maximizing energy efficiency wearable IoT applications while maintaining data accuracy. Overall, this thesis will greatly  advance the wearable IoT system orchestration for optimal system configurations.   </p>