Stimulus-driven preferential attention to threat can be modulated by goal-driven attention (Hahn & Gronlund, 2007;Van Dillen, Lakens, & Van den Bos, 2011;Vogt, De Houwer, Crombez, & Van Damme, 2013). However, it remains unclear how this goal-driven modulation affects specific attentional components implied in threat interference. We hypothesize that goal-driven modulation most strongly impacts delayed disengagement from threat. A spatial cueing task was used that disentangles delayed disengagement from attentional capture by tightly manipulating the locus of attention at the time of target onset. Different top-down goals were induced by instructing participants to identify bird/fish targets (Experiment 1) or spider/cat targets (Experiment 2) among animal non-targets. Delayed disengagement from a non-target spider was observed only when the spider was part of the target set, not when it was task-irrelevant. This corroborates evidence that threat stimuli do not necessarily override goal-driven attentional control and that extended processing of threatening distractors is not obligatory.Keywords: threat, fear, attention, goal-driven, stimulus-driven The Spider does 3The spider does not always win the fight for attention: Disengagement from threat is modulated by goal set.We are generally able to exert control over where we direct our attention.However, objects in the visual world will, at times, drive our attention. By its nature this stimulus-driven attention may act in opposition to the goal-driven (i.e. top-down) control of attention. An ongoing debate on the nature of attentional control has raised the question whether attention to salient distracter stimuli is involuntary and driven solely by stimulus properties such as abrupt onsets (Schreij, Theeuwes, & Olivers, 2010;Yantis & Jonides, 1984), color, and salience (e.g. Belopolsky, Schreij, & Theeuwes, 2010;Posner, 1980;Theeuwes, 1994;), or is modulated by task goals. For example, salient but irrelevant distractors in spatial cueing have been shown to capture attention only when they contained a target feature (Folk, Remington, & Johnston, 1992). Folk et al. showed that a single abrupt-onset distractor captured attention if the target was a single abrupt-onset stimulus, but not when the target was a color singleton. Conversely, a color singleton distractor captured attention if the target was a color singleton, but not when the target was a single stimulus defined by its abrupt onset. Consequently, it has been argued that attentional capture is contingent on an attentional control setting for task-relevant properties (contingent attentional capture; Folk et al.). A large number of subsequent studies (e.g. Dombrowe, Donk, & Olivers, 2011;Eimer, Kiss, Press, & Sauter, 2009;Folk, Remington, & Wu, 2009;Irons, Folk, & Remington, 2011;Wolfe, Butcher, Lee, & Hyle, 2003) have supported that one's attentional set modulates distractor interference, thus advocating top-down modulation of attentional interference.Against the background of the debate on the nature o...