Recent years have seen remarkable advances in the development and use of Artificial Intelligence (AI) in image classification, driving cars, and writing scientific articles. Although AI can outperform humans in many tasks, there remain domains where humans and AI working together can outperform either working alone. For humans and AI to work together effectively, the human must trust the AI bot to the right degree (calibrated). If the human does not trust the bot sufficiently, or conversely trusts the bot more than is warranted, the human-bot team will not perform as well as they could. We report three experiments examining trust in human-AI teaming. While existing studies typically collect binary responses (to trust, or not to trust), we present a novel paradigm that quantifies trust in a bot-recommendation in a continuous fashion. These data allow better precision, and in the future the development of more refined models of human-bot trust.

We present an analysis of real-life televised Dutch Auctions, where over the course of 12 hours of auctions, bidders made over 5,000 purchases of 81 unique products. In the Dutch Auction format, multiples of an item (traditionally Dutch tulips) are for sale, where prices start high and decrease over time. In the televised version, viewers of the Price-Drop TV program could wait for the price to drop over time but needed to weigh this against the risk of being scooped by the collective bidding behaviour of other viewers. We conduct analyses of the auctions to: 1) demonstrate the features of sell-out auctions where all items are sold, 2) show how time pressure increases sales in the auction, and 3) how the type of price decrease, discrete vs. continuous, affects people's decision to bid on items. We conclude by proposing an iterative process model based on the framework of Prospect Theory to extend theoretically motivated, laboratory-based findings out to real-world purchasing behaviours.

For the last 80 years, the Stroop task has been used totest theories of attention and cognitive control and ithas been applied in many clinical settings. Most theo-ries posit that the overwhelming power of written wordsovercomes strict instructions to focus on print color andignore the word. Recent evidence suggests that trials inthe Stroop task could in fact be a mixture of readingtrials and non-reading trials. Here we conduct a criticaltest of this mixture hypothesis, where a mixture of pro-cesses should satisfy the fixed-point property (Falmagne,1968).

In a Dutch auction, an item is offered for sale at a set maximum price. The price is then gradually lowered over a fixed interval of time until a bid is made, securing the item for the bidder at the current price. Bidders must trade-off between certainty and price: bid early to secure the item and you pay a premium; bid later at a lower price but risk losing to another bidder. These properties of Dutch auctions provide new opportunities to study competitive decision-making in a group setting. We developed a novel computerised Dutch auction platform and conducted a set of experiments manipulating volatility (fixed vs varied number of items for sale) and price reduction interval rate (step-rate). Triplets of participants ([Formula: see text]) competed with hypothetical funds against each other. We report null effects of step-rate and volatility on bidding behaviour. We developed a novel adaptation of prospect theory to account for group bidding behaviour by balancing certainty and subjective expected utility. We show the model is sensitive to variation in auction starting price and can predict the associated changes in group bid prices that were observed in our data.

Collaboration in shared environments requires human agents to coordinate their behaviour according to the machines’ actions. In this study, we compared the performance and behaviour of Human-Machine (HM) and Human-Human (HH) teams. While HH teaming behaviour is sensitive to Collaborative contexts, little is known about HM teaming behaviour. Furthermore, teaming behaviour may impact the team’s Joint Capacity the team’s ability to handle teamwork processes and task demands. To assess teaming behaviour at every moment of a trial we used three distinct spatiotemporal measures (Momentary Distance, Highly Correlated Segments, and Running Correlation). To assess the team's joint performance, we adopted the Capacity Coefficient (Townsend & Nozawa, 1995). For both HH and HM teams, behavioural measures predicted Joint Capacity. HH teams demonstrated greater performance and less synchronous behaviour than HM teams. The reduced synchrony of HH teams likely improved their performance as they could complement each other’s behaviour rather than duplicate inefficiencies.