Loot boxes are a purchasable video-game feature consisting of randomly determined, in-game virtual items. Due to their chance-based nature, there is much debate as to whether they constitute a form of gambling. We sought to address this issue by examining whether players treat virtual loot box rewards in a way that parallels established reward reactivity for monetary rewards in slots play. Across two sets of experiments, we show that loot boxes containing rarer items are more valuable, arousing, rewarding and urge-inducing to players, similar to the way slots gamblers treat rare large wins in slots play. Importantly, we show in Experiment 2 that the duration of Post Reinforcement Pauses, an index of reward reactivity, are longer for boxes with rarer items. Boxes containing rarer rewards also trigger larger Skin Conductance Responses and larger force responses—indices of positive arousal. Findings of Experiment 2 also revealed that there was an increase in anticipatory arousal prior to the reveal of loot box rewards. Collectively, our results elucidate the structural similarities between loot boxes and specific gambling games. The fact that players find rarer game items hedonically rewarding and motivating has implications for potential risky or excessive loot box use for some players.
We investigated whether anxious individuals, who adopt an inherently negative mindset, demonstrate a particularly salient memory bias for words tainted by negative contexts. To this end, sequentially presented target words, overlayed onto negative or neutral pictures, were studied in separate blocks (within-subjects) using a deep or shallow encoding instruction (between-subjects). Following study, in Test 1, participants completed separate recognition test blocks for the words overlayed onto the negative and the neutral contexts. Following this, in Test 2, participants completed a recognition test for the foils from each Test 1 block. We found a significant three-way interaction on Test 2, such that individuals with high anxiety who initially studied target words using a shallow encoding instruction, demonstrated significantly elevated memory for foils that were contained within the negative Test 1 block. Results show that during retrieval (Test 1), participants re-entered the mode of processing (negative or neutral) engaged at encoding, tainting the encoding of foils with that same mode of processing. The findings suggest that individuals with high relative to low anxiety, adopt a particularly salient negative retrieval mode, and this creates a downstream bias in encoding and subsequent retrieval of otherwise neutral information.