Backward Blocking Research Articles

Retrospective revaluation and its neural circuit in rats

Contingency learning is essential for establishing predictive or causal judgements. Retrospective revaluation captures essential aspects of the updating of this knowledge, according to new experience. In the present study, retrospective revaluation and its neural substrate was investigated in a rat conditioned magazine approach. One element of a previously food-reinforced Tone-Light compound stimulus was either further reinforced (inflation) or extinguished (extinction). These treatments affected the predictive value of the alternate stimulus (target), but only when the target was a weakly salient stimulus such as a Light, and the inflation/extinction procedure concerned the more salient element, that is the Tone. As the predictive value of the Light was decreased in comparison with a relevant control group, this revaluation was interpreted as backward blocking, and not unovershadowing. This observation challenges retrospective revaluation models focused on acquisition and prediction error detection, and is better accounted for by retrieval-based associative theories such as the comparator model (Miller and Matzel) [5]. Immunohistochemical detection of the Fos protein after the test phase revealed activation of the orbitofrontal and infralimbic cortices as well as nucleus accumbens core and shell, in rats that exhibited retrospective revaluation. Our results suggest that rats integrate successive experiences at the retrieval stage of retrospective revaluation, and that prefronto–accumbal interactions are involved in this function.

Backward Blocking and Interference Between Cues are Empirically Equivalent in Non-Causally Framed Learning Tasks

Backward blocking (BB) and interference between cues (IbC) are cue competition effects produced by very similar manipulations. In a standard BB design, both effects might occur simultaneously, which implies a potential problem for studying BB. In the present study with humans, the magnitude of both effects was compared using a non-causal scenario and a within-subjects design. Previous studies have made this comparison using learning tasks framed within causal scenarios. This posits a limit to generalizing their findings to non-causal learning situations because there is ample evidence showing that participants engage in causal reasoning when tasks are causally framed. The results obtained showed BB and IbC effects of the same magnitude in a non– causally-framed task. This highlights the methodological need for an IbC control in BB experiments.

A memory-based account of retrospective revaluation.

We adapt an instance model of human memory, Minerva 2, to simulate retrospective revaluation. In the account, memory preserves the events of individual trials in separate traces. A probe presented to memory contacts all traces in parallel and causes each to become active. The information retrieved from memory is the sum of the activated traces. Learning is modelled as a process of cued-recall; encoding is modelled as a process of differential encoding of unexpected features in the probe (i.e., expectancy-encoding). The model captures three examples of retrospective revaluation: backward blocking, recovery from blocking, and backward conditioned inhibition. The work integrates an understanding of human memory and complex associative learning.

"Backward blocking in first-order conditioning": Correction to Urushihara & Miller (2010).

"Backward blocking in first-order conditioning": Correction to Urushihara & Miller (2010).

Spontaneous recovery from interference between cues but not from backward blocking

In the present study, we examined the differential effect on backward blocking (BB) and on interference between cues (IbC) of including a delay right before the test phase vs. between training phases 1 and 2 in humans. While models of IbC predict a spontaneous recovery (SR) of responding if the delay is placed immediately before the test instead of between phases 1 and 2, BB models predict that no difference should be observed due to the position of the delay. In our experiment, we obtained the SR from IbC but not from BB. These results suggest that backward blocking and interference between cues are likely to be the result of different processes.

Backward blocking in first-order conditioning.

Three lick suppression experiments with rats investigated backward blocking in first-order conditioning. As has been suggested in prior studies, the experiments demonstrated that backward blocking is difficult to obtain in conventional first-order conditioning situations. However, the authors demonstrate here that backward blocking is observed in first-order conditioning if the target cue's behavioral control is weak at the time of elemental training of the blocking cue. The target cue's behavioral control was weakened through forward blocking of the target cue by a third cue (Experiment 1), conducting compound and elemental training with backward temporal relationships to the unconditioned stimulus (Experiment 2), and extinguishing the target cue following compound training (Experiment 3). The results of these experiments suggest that weak control of behavior by the blocked cue at the time of elemental training of the blocking cue is a critical determinant of whether blocking can be observed. Prior failures to detect backward blocking in first-order conditioning are seemingly due to a difficulty in decreasing the response-eliciting potential of a cue by indirect means such as associative inflation of a competing cue.

波长交换光网络中基于路径计算单元的冲突解决方案设计

A contention resolution scheme, P-cycle auto-switching (PCAS), is proposed in wavelength-switched optical networks (WSONs) based on generalized multi-protocol label switching (GMPLS) to solve backward blocking. This system permits the preferable request and switches the other to a backup path generated by the P-cycle. Simulation results show that the PCAS scheme is effective in mitigating backward blocking, especially under concentrated load.

Causal powers and preventers: An explanatory account of cue interaction effects in human causal judgement

According to the causal powers theory, all causal relations are understood in terms of causal powers of one thing producing an effect by acting on liability of another thing. Powers can vary in strength, and their operation also depends on the presence of preventers. When an effect occurs, there is a need to account for the occurrence by assigning sufficient strength to produce it to its possible causes. Contingency information is used to estimate strengths of powers and preventers and the extent to which they account for occurrences and nonoccurrences of the outcome. People make causal judgements from contingency information by processes of inference that interpret evidence in terms of this fundamental understanding. From this account it is possible to derive a computational model based on a common set of principles that involve estimating strengths, using these estimates to interpret ambiguous information, and integrating the resultant evidence in a weighted averaging model. It is shown that the model predicts cue interaction effects in human causal judgement, including forward and backward blocking, second and third order backward blocking, forward and backward conditioned inhibition, recovery from overshadowing, superlearning, and backward superlearning.

Cue competition effects and young children’s causal and counterfactual inferences.

The authors examined cue competition effects in young children using the blicket detector paradigm, in which objects are placed either singly or in pairs on a novel machine and children must judge which objects have the causal power to make the machine work. Cue competition effects were found in a 5- to 6-year-old group but not in a 4-year-old group. Equivalent levels of forward and backward blocking were found in the former group. Children's counterfactual judgments were subsequently examined by asking whether or not the machine would have gone off in the absence of 1 of 2 objects that had been placed on it as a pair. Cue competition effects were demonstrated only in 5- to 6-year-olds using this mode of assessing causal reasoning.

Label Preference Schemes for Lightpath Provisioning and Restoration in Distributed GMPLS Networks

In wavelength-routed networks based on a GMPLS control plane, the resource reservation protocol with traffic engineering extensions (RSVP-TE) allows to establish end-to-end lightpaths. The resource reservation can be blocked due to lack of available resources (forward blocking) or due to resource contentions (backward blocking). In wavelength-routed networks, the backward blocking is the predominant blocking contribution, when traffic load is low or highly-dynamic and when lightpath restoration takes place. To reduce the backward blocking, the paper proposes two label preference (LP) schemes compliant with RSVP-TE message exchanges. LP schemes provide the destination node with a label identifying the preferred wavelength to reserve. The preferred label is computed in a distributed way during the forward signaling phase, with the objective of assigning disjoint wavelengths to reservation attempts that may contend the resources. Simulation results demonstrate that, compared to other schemes, LP schemes are effective in reducing the backward blocking during both lightpath provisioning and restoration, without negatively impacting the forward blocking.

Three-Year-Olds’ Retrospective Revaluation in the Blicket Detector Task

We presented 3-year-olds with backward blocking and recovery from overshadowing contingencies in the blicket detector task, a causal induction task that uses binary, deterministic outcomes. Results revealed recovery from overshadowing but no backward blocking. These results are consistent with recent inferential and computational models of causal learning and induction. Our findings extend and clarify recent reports of retrospective revaluation in 3- and 4-year-olds (Sobel, D. M., Tenenbaum, J. B., & Gopnik, A. (2004). Children's causal inferences from indirect evidence: Backwards blocking and Bayesian reasoning in preschoolers. Cognitive Science, 28, 303-333), and underscore the sophistication of causal induction processes in young children.

Second-order retrospective revaluation in human contingency learning

We demonstrated second-order retrospective revaluation with three cues (T1, T2, and C) and an outcome, in human contingency learning. Experimental task, PC-controlled video game in which participants were required to observe about the relations between firing missiles and the tank destruction, consisted of three training phases and two rating phases. Groups C+ and C- consisted of same first two training phases, CT+ (cues C and T with an outcome) and T1T2+ followed by C+, or C- training for Groups C+, C-, respectively. In rating phases, it is clearly demonstrated that the judgment of predictive value for the outcome of the T2 were higher by C+ training (second-order unovershadowing) and lowered by C- training (second-order backward blocking). The results for Groups RC+ and RC-, in which the orders of the first two training phase for Groups C+ and C- were interchanged, also showed second-order unovershadowing and second-order backward blocking. These results, the robustness of second-order retrospective revaluation against the order of the first training phases, can be explained by the extended comparator hypothesis and probabilistic contrast model. However, these results cannot be explained by traditional associative learning models.

Bayesian approaches to associative learning: From passive to active learning

Traditional associationist models represent an organism's knowledge state by a single strength of association on each associative link. Bayesian models instead represent knowledge by a distribution of graded degrees of belief over a range of candidate hypotheses. Many traditional associationist models assume that the learner is passive, adjusting strengths of association only in reaction to stimuli delivered by the environment. Bayesian models, on the other hand, can describe how the learner should actively probe the environment to learn optimally. The first part of this article reviews two Bayesian accounts of backward blocking, a phenomenon that is challenging for many traditional theories. The broad Bayesian framework, in which these models reside, is also selectively reviewed. The second part focuses on two formalizations of optimal active learning: maximizing either the expected information gain or the probability gain. New analyses of optimal active learning by a Kalman filter and by a noisy-logic gate show that these two Bayesian models make different predictions for some environments. The Kalman filter predictions are disconfirmed in at least one case.

Impact of routing and wavelength selection strategies on GMPLS-controlled distributed restoration

The restorability problem in wavelength-routed networks without wavelength converters is investigated. A distributed control plane based on the generalized multiprotocol label switching (GMPLS) suite is considered for disseminating link-state information and reserving network resources. Different routing and wavelength selection strategies are proposed and a detailed performance analysis is carried out during restoration. Simulation results evaluate the blocking probability due to both lack of available resources (i.e., forward blocking) and resource contentions (i.e., backward blocking). Performance comparison provides some insightful suggestions to be applied to GMPLS-controlled wavelength-routed networks without wavelength converters. First, it is recommended to perform routing and wavelength selection separately at the lightpath source and destination, respectively. Second, routing can be computed on link-state information advertised during provisioning, without the need of flooding link-state information during restoration. In fact, during restoration, the resource contention among concurrent signaling instances is the dominant cause of blocking (i.e., backward blocking overcomes forward blocking). For this reason, intelligent routing and wavelength selection strategies aiming at avoiding or limiting resource contentions can help to improve restorability. In particular, among those strategies, the proposed wavelength selection has a better ability to reduce resource contentions than the routing strategies.

Short Article: Backward Blocking: The Role of Within-Compound Associations and Interference between Cues Trained Apart

Most theoretical accounts of backward blocking place heavy stress on the necessity of the target cue having been trained in compound with the competing cue to produce a decrement in responding. Yet, other evidence suggests that a similar reduction in responding to the target cue can be observed when the outcome is later paired with a novel cue never trained in compound with the target cue (interference between cues trained apart). The present experiment shows that pairing another nonassociated cue with the same outcome may be sufficient to produce a decremental effect on the target cue, but the presence of a within-compound association between the target and the competing cue adds to this effect. Thus, both interference between cues trained apart and within-compound associations independently contribute to backward blocking.

The dichotomy between associative learning and comparative cognitive science and a possible integration between the two approaches

Associative learning has contributed to the advancement of understanding animal cognition. This approach has proposed useful models of the underlying mechanisms in the acquisition of behaviors, whereas comparative cognitive science provides explanations on animal cognition from evolutionary perspectives. In this article, I first demonstrate evidence to the value of the associative learning mechanisms in the explanation of the blocking and backward blocking phenomena (retrospective revaluation) in both humans and nonhuman animals. Then, I review a study of episodic-like memory in western scrub jays in order to show methodological and theoretical linkages between associative learning and comparative cognitive science. Although levels of explanation in cognitive processes studied in the two approaches seem to be different, I conclude that associative learning can provide valuable simple accounts on various issues in studies of comparative cognitive science. The integration of the two approaches will be critical to the further advancement of studies of animal psychology.

A Contention Detection Scheme for Lightpath Restoration in GMPLS Networks

In wavelength-routed networks based on GMPLS control plane, lightpaths can be dynamically established upon distributed resource reservation performed by RSVP-TE protocol. Resource contentions may occur in the backward signaling when two or more RSVP-TE instances concurrently attempt to reserve the same wavelength (backward blocking). This paper proposes a contention detection (CD) scheme, compliant with RSVP-TE, that detects possible resource contentions and adapts the wavelength selection strategy accordingly. CD scheme is applied to restoration of the lightpaths disrupted by a single-link failure. Simulation results show that the CD scheme is effective in mitigating the backward blocking and thus the overall blocking.

Asymmetries in cue competition in forward and backward blocking designs: Further evidence for causal model theory

A hallmark feature of elemental associative learning theories is that multiple cues compete for associative strength when presented with an outcome. Cue competition effects have been observed in humans, both in forward and in backward blocking procedures (e.g., Shanks, 1985) and are often interpreted as evidence for an associative account of human causal learning (e.g., Shanks & Dickinson, 1987). Waldmann and Holyoak (1992), however, demonstrated that cue competition only occurs in predictive, and not diagnostic, learning paradigms. While unexplainable from an associative perspective, this asymmetry readily follows from structural considerations of causal model theory. In this paper, we show that causal models determine the extent of cue competition not only in forward but also in backward blocking designs. Implications for associative and inferential accounts of causal learning are discussed.

Prospective and Retrospective Learning in Honeybees

We focus on non-elemental forms of learning in honeybees in order to answer the question of whether retrospective learning can be found in an insect. We analyze three different forms of learning: category learning, rule learning and backward blocking. We provide examples showing that honeybees demonstrate these three forms of learning and propose that causal retrospection underlies them to different extents. We argue that an elemental associative account explains category learning whereas rule learning may require retrospection. Backward blocking, on the other hand, admits interpretations based on prospective learning. Consequently, because animals, including honeybees, solve these three types of problems, distinguishing between species on the basis of these capacities is inappropriate.

Forward blocking in human learning sometimes reflects the failure to encode a cue–outcome relationship

In two "allergist" causal judgement experiments, participants were trained with a blocking design (A+|AB+). The procedure allowed different food cues to be paired with different fictitious allergic reactions. On test, participants were asked to rate the causal efficacy of the target cues and to recall the particular allergic reaction (outcome) that had followed each cue during training. Forward blocking was observed on the causal judgement measure and on the outcome recall measure in both Experiment 1 and Experiment 2. A backward blocking contingency was also trained in Experiment 2 (AB+|A+). Backward blocking was not observed either on the causal judgement or on the outcome recall measure. The evidence from the recall measure suggests that forward blocking in this task results from a failure to encode the B-outcome relationship during training. Associative and nonassociative mechanisms of forward blocking are discussed.