Feed-forward neural networks using cerebral MR spectroscopy and DTI might predict neurodevelopmental outcome in preterm neonates.

OBJECTIVES: We aimed to evaluate the ability of feed-forward neural networks (fNNs) to predict the neurodevelopmental outcome (NDO) of very preterm neonates (VPIs) at 12 months corrected age by using biomarkers of cerebral MR proton spectroscopy (1H-MRS) and diffusion tensor imaging (DTI) at term-equivalent age (TEA). METHODS: In this prospective study, 300 VPIs born before 32 gestational weeks received an MRI scan at TEA between September 2013 and December 2017. Due to missing or poor-quality spectroscopy data and missing neurodevelopmental tests, 173 VPIs were excluded. Data sets consisting of 103 and 115 VPIs were considered for prediction of motor and cognitive developmental delay, respectively. Five metabolite ratios and two DTI characteristics in six different areas of the brain were evaluated. A feature selection algorithm was developed for receiving a subset of characteristics prevalent for the VPIs with a developmental delay. Finally, the predictors were constructed employing multiple fNNs and fourfold cross-validation. RESULTS: By employing the constructed fNN predictors, we were able to predict cognitive delays of VPIs with 85.7% sensitivity, 100% specificity, 100% positive predictive value (PPV) and 99.1% negative predictive value (NPV). For the prediction of motor delay, we achieved a sensitivity of 76.9%, a specificity of 98.9%, a PPV of 90.9% and an NPV of 96.7%. CONCLUSION: FNNs might be able to predict motor and cognitive development of VPIs at 12 months corrected age when employing biomarkers of cerebral 1H-MRS and DTI quantified at TEA. KEY POINTS: • A feed-forward neuronal network is a promising tool for outcome prediction in premature infants. • Cerebral proton magnetic resonance spectroscopy and diffusion tensor imaging can be used for the construction of early prognostic biomarkers. • Premature infants that would most benefit from early intervention services can be spotted at the time of optimal neuroplasticity.

When all is revealed: a dissociation between evaluative learning and contingency awareness.

Three experiments are reported that address the issue of awareness in evaluative learning in two different sensory modalities: visual and haptic. Attempts were made to manipulate the degree of awareness through a reduction technique (by use of a distractor task in Experiments 1 and 2 and by subliminally presenting affective stimuli in Experiment 3) and an induction technique (by unveiling the evaluative learning effect and requiring participants to try to discount the influence of the affective stimuli). The results indicate overall that evaluative learning was successful in the awareness-reduction groups but not in the awareness-induction groups. Moreover, an effect in the opposite direction to that normally observed in evaluative learning emerged in participants aware of the stimulus contingencies. In addition, individual differences in psychological reactance were found to be implicated in the strength and direction of the effect. It is argued that these results pose serious problems for the contention that awareness is necessary for evaluative learning.

Transfer effects as a function of sequential and quantitative schedule constraints.

Schwartz (1982, 1988) found that a pretraining of contingent reinforcement interferes with subsequent rule discovery. The present study investigated the effects of schedule imposed sequential and quantitative constraints (Timberlake & Allison, 1974) on task performance in a subsequent test phase. Sixty-four Ss, students of the University of Duesseldorf, were assigned at random to one of four experimental conditions, differing according to the presence vs. absence of sequential and quantitative constraints, respectively. Discrimination-learning performance and variability during test phase were significantly better for Ss experiencing sequential constraint during treatment. In contrast, the introduction of a quantitative restriction during treatment had no statistically significant effects on test phase performance.

Human evaluative conditioning: order of stimulus presentation.

In an experiment designed to demonstrate evaluative conditioning, subjects were shown 48 pictures of sculptures that they rated on a scale with 21 categories (-10 to +10). Then, the two most liked pictures (L) were paired with pictures from the categories -1, 0, or +1 (N). In contrast to prior experiments, subjects were given either forward conditioning (N-L) or backward conditioning (L-N) trials but not both. Four other neutral stimuli were paired with each other (N-N) and acted as control stimulus pairs. After conditioning, the stimuli were rated a second time. There was a statistically significant difference in evaluative ratings showing a change of the evaluative tone of the previously neutral stimuli in a positive direction only after forward conditioning. This finding is inconsistent with results of prior experiments and challenges the assumption of Martin and Levey (1987) that evaluative conditioning is different from human classical conditioning.