Development of a Selective Extinction through Cognitive Evaluation Paradigm

In this study we trialled a novel paradigm that aims to bridge the gap between the research into the extinction of conditioned responses and the instructed regulation of emotion via a prescribed strategy. This was done through a process we called Selective Extinction through Cognitive Evaluation (SECE). Participants were conditioned to associate a previously neutral stimulus (letter, CS+) with an aversive burst of white noise (US) and another letter with no aversive event (CS-). In a second phase, the same letters were presented within words belonging to two distinct categories. One group did not receive any more noise bursts during this phase which lead to the extinction of the conditioned responses (extinction group). The other group (SECE group) was told that the CS+ now only carried a risk of the noise occurring when it was presented in one (dangerous), but not the other word category (safe). Affective responses were assessed using skin conductance responses (SCR). We found overall increased SCR’s in response to CS+ (safe and dangerous) compared to CS- trials in the SECE group, and also increased SCR’s in response to dangerous compared to safe CS+ trials. This suggests that the conditioned response (CR) can be reduced through the cognitive evaluation of additional information and it is possible that extinction mechanisms are involved in this process. We did not find any effects in the extinction group. It is possible that extinction was completed too quickly so that any differences between conditions could not be captured. We therefore conducted a second study to pilot an improved paradigm. We visually amended the trials, amended the reinforcement schedule, and changed to a fully within-subjects design. Here we once again found significantly reduced SCR to safe compared to dangerous CS+ trials. Future experiments will use these results as a basis for an improved paradigm to better capture changes in SCR as well as BOLD fMRI in future experiments.

Classification of crystal structure using a convolutional neural network

A deep machine-learning technique based on a convolutional neural network (CNN) is introduced. It has been used for the classification of powder X-ray diffraction (XRD) patterns in terms of crystal system, extinction group and space group. About 150 000 powder XRD patterns were collected and used as input for the CNN with no handcrafted engineering involved, and thereby an appropriate CNN architecture was obtained that allowed determination of the crystal system, extinction group and space group. In sharp contrast with the traditional use of powder XRD pattern analysis, the CNN never treats powder XRD patterns as a deconvoluted and discrete peak position or as intensity data, but instead the XRD patterns are regarded as nothing but a pattern similar to a picture. The CNN interprets features that humans cannot recognize in a powder XRD pattern. As a result, accuracy levels of 81.14, 83.83 and 94.99% were achieved for the space-group, extinction-group and crystal-system classifications, respectively. The well trained CNN was then used for symmetry identification of unknown novel inorganic compounds.

Advances in space-group determination from powder diffraction data

The space-group determination process by powder diffraction data is not straightforward. The low accuracy of the reflection intensities may invalidate the calculation of the probability associated to each extinction group that is compatible with the crystal system determined in the indexation step. Here the combination of thezstatistics with two new algorithms is reported: the first checks the quality of each 2θ interval in order to omit doubtfulzestimates from the calculations; the second creates a list of reflections with peaks that weakly overlap with any other peak, in order to check if any of them violates the extinction rules of the extinction symbol. The new approach has been applied to a large set of test structures and proved to be much more efficient than the procedure based only on thezstatistics.

Teaching Children to Discriminate Letters of the Alphabet through Errorless Discrimination Training

A group of 64 inner city preschool children was taught to discriminate letters of the alphabet using either an errorless discrimination training (EDT) approach or the traditional reinforcement-extinction approach. For the EDT group the distinctive feature of the letter to be discriminated was highlighted in red and as the training proceeded the red was gradually faded. Each group received ten training trials for each of two different letter combinations presented in a match-to-sample format. The results showed that the EDT group made significantly fewer errors during training and on the posttest compared to the reinforcement-extinction group.

Delayed Reward, Delay-Box Confinement, and Instrumental Performance: A within-Type Design

Three experiments were done on delayed reinforcement and extinction using a within-group design in training. Entering one alley of a Grice discrimination apparatus led to immediate reward, while entering the other led to 40-sec. delays either on 50% of the trials (Exps. I and II) or on 100% of the trials (Exp. III). Half of Ss were extinguished in the non-delay alley, half in the delay alley. Half of Ss in each extinction group were extinguished without delay-box confinement, half with 40-sec. delay-box confinement on 50% of the trials (Exps. I and II) or 100% of the trials (Exp. III). In all three experiments Ss learned to discriminate between the delay and non-delay alleys. In Exp. I Ss ran significantly faster in extinction in the non-delay alley. The same trend was evident in Exps. II and III. Thus, the results were in the opposite direction to those which have been found in between-group studies. In Exp. III Ss with 100% delay-box confinement in extinction ran slower (.10 > p > .05) than Ss without confinement.

Effect of Number of Goal Box Cues on ‘Latent Extinction’

Rate of extinction and number of food cup responses were measured in rats given varying numbers of goal box cues during pre-extinction. Forty-five food-deprived rats were given equal acquisition in a T-maze with a clicking feeder and divided randomly into three groups according to number of goal box cues presented during pre-extinction (Group I, neutral box; Group II, goal box; Group III, goal box plus clicks). Analysis of variance of correct turns and running time data revealed that Group II Ss extinguished faster than Group I and III Ss, and Groups I and III did not differ significantly. These results suggested that strong goal box stimuli (clicks) have not only the capacity to elicit fractional anticipatory goal responses, but also an excitatory (reinforcing) capacity functioning independently of the rg eliciting capacity.