To understand how correlated the human brain is across multiple subjects or multiple trials when people make risky decisions, economical decision-making based on the Newsvendor Problems (NP) were investigated with 64-channel electro-encephalography (EEG) taken from 23 human subjects. Specifically, in this work, we analyzed the EEG signals recorded while subjects were playing NP decision-making tasks. Two groups of subjects were assigned to take NP with distinctive difficulty levels for examining the difference in neural responses to the task challenge. The Correlated Components Analysis (CorrCA) method was then utilized to identify the EEG channels that maximized the correlations among multiple subjects or trials. Both Inter-Subject Correlation (ISC) and Inter-Trial Correlation (ITC) results revealed an apparent modulation of the attentional state of subjects across three test phases (decision-making, feedback, and recovery) and two distinct task challenges. The neural correlation across multiple subjects or multiple trials increased significantly during the decision-making phase for more challenging tasks, in which subjects needed to be more attentive. Such observations confirm the modulation of the engagement state on the correlations of the neural activity across multiple subjects or repeated trials. Moreover, the alpha-band power results for different decision-making phases affirmed that the difficulty of different test phases modulates subjects' engagement.

A neural network based on fuzzy trace theory and previously used to model decisions Uy a retailer on how much of a product to purchase (the “newsvendor problem is extended to a case with a particularly salient gist. This is the case where the newsvendor is a manager who receives a bonus if profit is at least at a target value. The gist is that in order to maximize the probability of a Uonus the newsvendor should Uuy exactly the minimum amount that makes the target profit possible, that is, target divided Uy (price per item minus cost per item). The network includes amygdala and orUitofrontal cortex for emotional evaluation, anterior cingulate for Uehavioral planning, and striatum, thalamus, and motor cortex for Uehavioral execution.

A brain-based neural network previously used to model short-term decisions from description is extended to decisions from experience - specifically, decisions by a retailer on how much of a product to purchase per day based on cost, price, discounts for large purchases, and feedback about demand on previous days. Learning from feedback is represented in the network by an analog of dopamine neurons' computation of reward prediction error. This is superimposed on a network that includes amygdala and orbitofrontal cortex for emotional evaluation, anterior cingulate for behavioral planning, and striatum, thalamus, and motor cortex for behavioral execution.

In recent years, user-created content websites have gained extensive popularity. An example is provided by the HP Labs' website, MagCloud, a web service that enables anyone to publish his or her own magazine. In this paper, we explain how this novel platform works and how we can use a quantitative model to estimate the patterns of its adoption by publishers and consumers so far. The resulting multisided ecosystem captures non-trivial externalities between publishers, subscribers and potentially advertisers. Real world data was used to estimate the model and the results can be used to optimize marketing strategies to accelerate adoption as well as to forecast growth trajectories and opportunities. As a result of this work, we found that website visitors from different sources adopt its services with varying propensities. Our modeling framework allows us to identify whether increases in publications or orders originate from increases in website traffic or from past actions of publishers and consumers.

We experimentally compare low-information, high-information and self-reporting reputation mechanisms. The results indicate players strategically reacted to the reputation mechanisms, with higher information mechanisms increasing market efficiency.

We present a strategy for automatic negotiation that takes the same approach as computer programs that play games such as chess; we build the game tree. For every offer we look at every counteroffer, every counteroffer to each of them, and so on. The strategy then selects the counteroffer that has the largest expected payoff. A number of problems arise that are unique to using this strategy for negotiation. These include uncertainty in the opponent’s goals, the fact that a bad move can penalize both players, and moves that are continuous, as opposed to discrete. We show how the standard methods of building the tree and evaluating the results were adapted to this environment.

No abstract available.

We present a novel methodology for predicting future outcomes that uses small numbers of individuals participating in an imperfect information market. By determining their risk attitudes and performing a nonlinear aggregation of their predictions, we are able to assess the probability of the future outcome of an uncertain event and compare it to both the objective probability of its occurrence and the performance of the market as a whole. Experiments show that this nonlinear aggregation mechanism vastly outperforms both the imperfect market and the best of the participants.