We report on the optical spectroscopy of the eclipsing halo low-mass X-ray binary 2S 0921Ϫ630, which reveals the absorption-line radial velocity curve of the K0 III secondary star with a semiamplitude km K p 92.89 ‫ע‬ 3.84 2 s Ϫ1 , a systemic velocity km s Ϫ1 , and an orbital period of days (1 j). Given g p 34.9 ‫ע‬ 3.3 P 9.0035 ‫ע‬ 0.0029 orb the quality of the data, we find no evidence for the effects of X-ray irradiation. Using the previously determined rotational broadening of the mass donor and applying conservative limits on the orbital inclination, we constrain the compact object mass to be 2.0-4.3 M , (1 j), ruling out a canonical neutron star at the 99% level. Since the nature of the compact object is unclear, this mass range implies that the compact object is either a low-mass black hole with a mass slightly higher than the maximum possible neutron star mass (2.9 M , ) or a massive neutron star. If the compact object is a black hole, it confirms the prediction of the existence of low-mass black holes, while if the object is a massive neutron star, its high mass severely constrains the equation of state of nuclear matter.

The relationship between predictive learning and attentional processing was investigated in two experiments. During a learning procedure participants viewed rapid serial visual presentation (RSVP) of stimuli in the context of a choice-reaction-time (CRT) task. Salient stimuli in the RSVP streams were either predictive or non-predictive for the outcome of the CRT task.Following this procedure we measured attentional blink (AB) to the predictive and nonpredictive stimuli. In Experiment 1, despite the use of a large sample and checks demonstrating the validity of the learning procedure and the AB measure, we did not observe reduced AB for predictive stimuli. In contrast, in Experiment 2, where the predictive stimuli occurred alongside salient non-predictive comparison stimuli, we did find less AB for predictive than for nonpredictive stimuli. Our results support an attentional model of learning in which relative prediction error is used to increase learning rates for good predictors and reduce learning rates for poor predictors (Mackintosh, 1975) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Attentional Blink 3 Relative prediction error and protection from attentional blink in human associative learning Several models of learning suggest that learning, a central process of association formation, is accompanied by changes in attention to task relevant cues. However, models do not usually characterize what attention entails beyond a change in a learning rate parameter in a mathematical model (e.g. Le Pelley, 2004;Mackintosh, 1975;Pearce & Hall, 1980). The current paper follows-up a study by Livesey, Harris, and Harris (2009) which showed reduced attentional blink (AB) to stimuli that had predictive value in a choice-reaction-time (CRT) task.This AB learning effect demonstrated a correspondence between AB and the learning rate changes described in Mackintosh's (1975) model of associative learning, making concrete a link between an abstract mathematical model and a psychological process. Within the associative learning literature, Mackintosh's (1975) model has frequently been considered incompatible with Pearce and Hall's (1980) model. Both models make use of the concept of prediction error in terms of the difference between the outcome of a learning trial and an expectation derived from the associative strength of the conditioned stimuli (CSs) that are present on that trial. On each trial, the Mackintosh model adjusts learning rates according to relative prediction error whereas the Pearce-Hall model makes use of absolute prediction error. In the Mackintosh model, the learning rate for CSs that are best predictors within a learning task increases, hence these are considered to receive more attention. In contrast, in the Pearce-Hall model, good predictors have their learning rates reduced, hence are considered to receive less atte...