Conditional Discrimination Research Articles

Peripheral vision in matching-to-sample procedures.

Eye-tracking has been used to investigate observing responses in matching-to-sample procedures. However, in visual search, peripheral vision plays an important role. Therefore, three experiments were conducted to investigate the extent to which adult participants can discriminate stimuli that vary in size and position in the periphery. Experiment 1 used arbitrary matching with abstract stimuli, Experiment 2 used identity matching with abstract stimuli, and Experiment 3 used identity matching with simple (familiar) shapes. In all three experiments, participants were taught eight conditional discriminations establishing four 3-member classes of stimuli. Four different stimulus sizes and three different stimulus positions were manipulated in the 12 peripheral test phases. In these test trials, participants had to fixate their gaze on the sample stimulus in the middle of the screen while selecting a comparison stimulus. Eye movements were measured with a head-mounted eye-tracker during both training and testing. Experiment 1 shows that participants can discriminate small abstract stimuli that are arbitrarily related in the periphery. Experiment 2 shows that matching identical stimuli does not affect discrimination in the periphery compared to arbitrarily related stimuli. However, Experiment 3 shows that discrimination increases when stimuli are well-known simple shapes.

Reducing the occurrence of mouthing and jumping in a dog through conditional discrimination training.

Many approaches for reducing unwanted behavior use punishment, extinction, or noncontingent reinforcement. Other methods focus on teaching and reinforcing alternative behaviors that can replace the unwanted behavior. Another strategy can be to change the stimulus control of the unwanted behavior. The present study investigated if conditional discrimination training using positive reinforcement could reduce undesirable behaviors in a pet dog. After conditional discrimination training, two unwanted behaviors (jumping and mouthing) occurred reliably in the presence of new discriminative stimuli, while other behaviors occurred in the presence of the discriminative stimuli that had previously produced these unwanted behaviors. This experiment demonstrates that conditional discrimination training can be an effective way to control the frequency of an unwanted behavior by controlling the frequency of the presentation of a discriminative stimulus.

Discrimination of clinical and nonclinical conditions of the digital application S-ONapp for the assessment and testing of sexual dysfunctions

Objective: The current study aims to evaluate the diagnostic validity of the scales included in the S-ONapp application. Methods: This cross-sectional study was conducted between February 28, 2022, and April 30, 2022, on subjects from the USA, Europe, Canada and Asia. A clinical group of 430 subjects and a non-clinical control group of 420 subjects were drawn. Results: Multiple comparisons were made between the two groups, the S-ONapp application proving its discriminatory capacity between the clinical and non-clinical groups. Conclusions: The S-ONapp application, includes sexual dysfunction assessment and testing tools that present a clinical robustness in the diagnosis of sexual dysfunctions.

Golden-Sine dynamic marine predator algorithm for addressing engineering design optimization

In engineering design optimization problems, the optimal solution can improve the design quality of complex engineering system and reduce a lot of cost consumption, so it is of great practical significance to study the optimization algorithm of engineering design problems. Evolutionary computation is widely used to solve engineering design optimization problems, which are mostly mixed-integer nonlinear programming (MINLP) problems. As a newly developed evolutionary computing method, Marine Predator Algorithm (MPA) currently suffers from weak convergence and easily falls into local optimum. In order to overcome the disadvantage, this study proposed a Golden-Sine Dynamic Marine Predator Algorithm (GDMPA). Firstly, Logistic-Logistic (L-L) cascade chaos was used to adjust the initial position of the population to generate a high-quality initial prey population while ensuring ergodicity and randomness. Secondly, the dynamic adjustment transition probability strategy was added to improve the discriminant conditions when predators entered different stages, which effectively maintained the balance between global exploration and local exploitation. The adaptive inertial weight based on Sigmoid function was used in updating the step information of predators to avoid the problem of falling into local extrema. Finally, the Golden-Sine factor is employed to achieve a better balance between exploration and exploitation, further improve the premature convergence problem, enhance the population diversity, and improve the convergence rate. A series of validation studies were conducted over twelve standard test functions and the CEC2017 test set to verify the effectiveness and robustness of the improved GDMPA strategy. Mechanical optimization and size optimization study for truss structures was carried out using the proposed GDMPA, which yielded excellent results. The results of the 27-bar truss structure show that the proposed GDMPA reduces 3.24%, 27.18%, 39.38%, 27.65% and 9.67% compared to the total mass of MPA, BOA, SSA, SOA and HHO, respectively. In the other cases, the optimization results of GDMPA have been improved substantially compared with other algorithms. Therefore, GDMPA has a broad application prospect in structural design and optimization.

Passive-active fault detection of Boolean control networks

This paper investigates the fault detectability of Boolean control networks. To deal with different circumstances, passive fault detectability and four types of active fault detectability are proposed for the first time. For passive fault detectability, in line with observed input-output data, several discriminant conditions are put forward. For active fault detectability, by resorting to an auxiliary system, the essence of active fault detectabilty is revealed via reachability. Subsequently, corresponding to different types of active fault detectability, algebraic criteria that facilitate verification are provided. At the same time, the relationships of four types of active fault detectability are discussed as well. In addition, an approach for calculating fault detection time is given by combining passive fault detectability with active fault detectability. Finally, the effectiveness of the results obtained is elaborated by several examples.

Flexible learning of matching and mismatching by pigeons

When pigeons learn a conditional discrimination in which a sample stimulus indicates which of two comparison stimuli is correct Skinner (1950) proposed that they learn a chain involving the sample, the correct comparison stimulus, and the reinforcer. This implies that they do not learn to reject the incorrect comparison stimulus and the sameness relation between the sample and the correct or the incorrect comparison stimulus plays little role in learning. There is, however, considerable evidence that learning to match or mismatch the sample can transfer to novel stimuli. But there is little evidence that the sameness relation facilitates acquisition. In the present research, pigeons were trained on two 0-s delay conditional discriminations: two matching tasks, two mismatching tasks, or one of each. No differences were found in acquisition, suggesting that consistent matching or mismatching does not facilitate acquisition of conditional discriminations. In testing, when either the correct or the incorrect comparison stimulus in each discrimination was replaced with one of the stimuli from the other task, results suggested that when learning both tasks, the pigeons in all three groups had learned both to select the correct stimulus and to reject the incorrect stimulus. It appears that the pigeons may have had learned the tasks based on sample/comparison-stimulus configurations with the sample serving as an occasion setter.

Necessity and recruitment of cue-specific neuronal ensembles within the basolateral amygdala during appetitive reversal learning

Through Pavlovian appetitive conditioning, environmental cues can become predictors of food availability. Over time, however, the food, and thus the value of the associated cues, can change based on environmental variations. This change in outcome necessitates updating of the value of the cue to appropriately alter behavioral responses to these cues. The basolateral amygdala (BLA) is critical in updating the outcomes of learned cues. However, it is unknown if the same BLA neuronal ensembles that are recruited in the initial associative memory are required when the new cue-outcome association is formed during reversal learning. The current study used the Daun02 inactivation method that enables selective targeting and disruption of activated neuronal ensembles in Fos-lacZ transgenic rats. Rats were implanted with bilateral cannulas that target the BLA and underwent appetitive discriminative conditioning in which rats had to discriminate between two auditory stimuli. One stimulus (CS+) co-terminated with food delivery, and the other stimulus was unrewarded (CS−; counterbalanced). Rats were then tested for CS+ or CS− memory retrieval and infused with either Daun02 or a vehicle solution into the BLA to inactivate either CS+ or CS− neuronal ensembles that were activated during that test. To assess if the same neuronal ensembles are necessary to update the value of the new association when the outcomes are changed, rats underwent reversal learning: the CS+ was no longer followed by food (reversal CS−, rCS−), and the CS− was now followed by food (reversal CS+; rCS+). The group that received Daun02 following CS+ session showed a decrease in conditioned responding and increased latency to the rCS− (previously CS+) during the first session of reversal learning, specifically during the first trial. This indicates that the neuronal ensemble that was activated during the recall of the CS+ memory was the same neuronal ensemble needed for learning the new outcome of the same CS, now rCS−. Additionally, the group that received Daun02 following CS− session was slower to respond to the rCS+ (previously CS−) during reversal learning. This indicates that the neuronal ensemble that was activated during the recall of the CS− memory was the same neuronal ensemble needed for learning the new outcome of the same CS. These results demonstrate that different neuronal ensembles within the BLA mediate memory recall of CS+ and CS− cues and reactivation of each cue-specific neuronal ensemble is necessary to update the value of that specific cue to respond appropriately during reversal learning. These results also indicate substantial plasticity within the BLA for behavioral flexibility as both groups eventually showed similar terminal levels of reversal learning.

Identifying Language Disorder in Bilingual Children Using Automatic Speech Recognition.

The differential diagnosis of developmental language disorder (DLD) in bilingual children represents a unique challenge due to their distributed language exposure and knowledge. The current evidence indicates that dual-language testing yields the most accurate classification of DLD among bilinguals, but there are limited personnel and resources to support this practice. The purpose of this study was therefore to determine the feasibility of dual-language automatic speech recognition (ASR) for identifying DLD in bilingual children. Eighty-four Spanish-English bilingual second graders with (n = 25) and without (n = 59) confirmed diagnoses of DLD completed the Bilingual English-Spanish Assessment-Middle Extension Morphosyntax in both languages. Their responses on a subset of items were scored manually by human examiners and programmatically by a researcher-developed ASR application employing a commercial speech-to-text algorithm. Results demonstrated moderate overall item-by-item scoring agreement (k = .54) and similar classification accuracy values (human = 92%, ASR = 88%) between the two methods using the best-language score. Classification accuracy of the ASR method increased to 94% of cases correctly classified when test items with poorer discrimination in the ASR condition were eliminated. This study provides preliminary support for the technical feasibility of ASR as a bilingual expressive language assessment tool. https://doi.org/10.23641/asha.20249994.

Comparison of mastery criteria applied to individual targets and stimulus sets on acquisition of tacts, intraverbals, and listener responses.

Mastery criteria can be applied to individual targets or stimuli organized into sets. Wong et al. (2021) and Wong and Fienup (2022) found that participants who received special education services learned sight words more rapidly when an individual target mastery criterion was applied. The current study replicated and extended these findings across novel skills. Five participants with ASD received tact or intraverbal training in Experiment 1, and 2 participants with ASD received auditory-visual conditional discrimination training (AVCD) in Experiment 2. In both experiments, mastery criteria were applied to targets and stimulus sets to compare sessions to mastery. Results showed the target mastery criterion required fewer sessions of tact training for 3 of 5 participants and AVCD training for both participants. However, overselection of stimuli occurred for 20% of AVCD mastered targets, suggesting a false positive for acquisition of those targets. Maintenance was similar across conditions and experiments.

Label enhancement-based feature selection via fuzzy neighborhood discrimination index

As an essential preprocessing step, the feature selection for multi-label classification is an efficient tool to solve the high-dimensionality in training data under multiple semantics. However, in multi-label learning, the assumption of a uniform distribution is not practicable for some real applications where labels are related to the instance with different relative importance degrees. Motivated by this, the label enhancement algorithm is utilized to transform the logical values in multiple labels into label distributions with real values. To maintain the supervised information of label distributions, the discrimination index is combined with granular theory to describe the vagueness in feature and label spaces, and the fuzzy neighborhood discrimination index is presented to assess the discrimination information in features with continuous values. Furthermore, an innovative label enhancement-based feature selection approach is designed to handle the challenge caused by high dimensionality with multiple semantics. In this algorithm, the fuzzy conditional discrimination index is proposed by combining the fuzzy neighborhood discrimination index under the feature subset with the label similarity matrix of label distributions, and the significance of the feature is measured by the fuzzy conditional discrimination index, which is utilized to assess the discernibility information of labels under the feature. Furthermore, to validate the efficacy of our algorithm, a collection of experiments are conducted with five representative multi-label feature selection methods under twelve benchmark datasets. With six widespread evaluation indicators of multi-label classification, the experimental results indicate that our algorithm obtains superior performance against other algorithms.

Why Does Regulation (EC) No 1/2003 Provide for the Imposition of Penalties Only on Undertakings?: A Historical Perspective

According to Article 23(2)(a) of Regulation (EC) No 1/2003, the Commission may impose any fines for the breach of substantive competition rules only on ‘undertakings’. The provision does not contemplate the imposition of fines on individuals considered responsible for the conduct of the undertaking concerned, even though, as explained in this article, this position is not imposed by the relevant legal basis (Article 83 EC [the Treaty establishing the European Community in the version following the Treaty of Amsterdam], corresponding to Article 87 EEC Treaty [the Treaty establishing the European Economic Community in its version of 1957] and the present Article 103 TFEU [the Treaty on the Functioning of the European Union]) That legal basis would indeed allow such liability to be established. From a policy point of view, however, it is often argued that sanctions should also be imposed on such individuals. This raises the question as to why the Union legislator chose not to provide for corresponding powers on the Commission’s part. From the relevant legislative history exposed in this article, it emerges that the approach of Article 23(2)(a) ultimately originates from the ECSC Treaty [theTreaty establishing the European Coal and Steel Community]. In this latter context, it reflected the particular limits placed by its authors on that Treaty. It did not reflect the contemporary orientations at national level regarding punitive sanctions. The approach gradually found its way into Regulation (EC) No 1/2003 without any further specific discussion on the way, despite the different context offered by the relevant provisions of the EEC/EC Treaty. It was first transplanted into Regulation No 11, based on Article 79 EEC Treaty (today Article 95 TFEU) and concerning the abolition of discrimination in transport rates and conditions. The negotiations preceding the adoption of this Regulation brought to light strong reserves regarding the conferral of sanctioning powers upon the (then) EEC, and it was no doubt unrealistic to expect that it would have been granted powers more far-reaching than those available under the ECSC Treaty. When Regulation No 17 (and specifically Article 15) was elaborated, the drafters were aware of the scheme established by Regulation No 11, which served as a reference. No genuine discussion regarding possible powers to punish the persons considered responsible for the conduct of the ‘undertaking’ took place. It is likely that any such innovation would not have been realistic at that point. Lastly, the reform brought about by Regulation (EC) No 1/2003 focused on other aspects regarding the implementation of competition law, and the scheme of Article 15(2) of Regulation No 17 has been taken over into Article 23(2) of Regulation (EC) No 1/2013 without major changes. No consideration was given to the possible imposition of penalties on natural persons implicated in the commission of breaches of competition law. EU antitrust sanctions, European Commission, individual sanctions, Regulation No. 1/ 2003, Regulation No 17, Regulation No 11, ECSC Treaty, history

Controllable and Identity-Aware Facial Attribute Transformation.

Modifying facial attributes without the paired dataset proves to be a challenging task. Previously, approaches either required supervision from a ground-truth transformed image or required training a separate model for mapping every pair of attributes. These limit the scalability of the models to accommodate a larger set of attributes since the number of models that we need to train grows exponentially large. Another major drawback of the previous approaches is the unintentional gain of the identity of the person as they transform the facial attributes. We propose a method that allows for controllable and identity-aware transformations across multiple facial attributes using only a single model. Our approach is to train a generative adversarial network (GAN) with a multitask conditional discriminator that recognizes the identity of the face, distinguishes real images from fake, as well as identifies facial attributes present in an image. This guides the generator into producing an output that is realistic while preserving the person's identity and facial attributes. Through this framework, our model also learns meaningful image representations in a lower dimensional latent space and semantically associate separate parts of the encoded vector with both the person's identity and facial attributes. This opens up the possibility of generating new faces and other transformations such as making the face thinner or chubbier. Furthermore, our model only encodes the image once and allows for multiple transformations using the encoded vector. This allows for faster transformations since it does not need to reprocess the entire image for every transformation. We show the effectiveness of our proposed method through both qualitative and quantitative evaluations, such as ablative studies, visual inspection, and face verification. Competitive results are achieved compared to the main competition (CycleGAN), however, at great space and extensibility gain by using a single model.

Fault Diagnosis under Varying Working Conditions with Domain Adversarial Capsule Networks

<p>Most existing studies that develop fault diagnosis methods focus on performance under steady operation while overlooking adaptability under varying working conditions. This results in the low generalization of the fault diagnosis methods. In this study, a novel deep transfer learning architecture is proposed for fault diagnosis under varying working conditions. A modified capsule network is developed by combining the domain adversarial framework and classical capsule network to simultaneously recognize the machinery fault and working conditions. The novelty of the proposed architecture mainly lies in the integration of the domain adversarial mechanism and capsule network. The idea of the domain adversarial mechanism is exploited in transfer learning, which can achieve a promising performance in cross-condition fault diagnosis tasks. With the novel architecture, learned features exhibit identical or very similar distributions in the source and target domains. Hence, the deep learning architecture trained in one working condition can be applicable to discriminative conditions without being hindered by the shift between the two domains. The proposed method is applied to analyze vibrations of a bearing system acquired under different working conditions, i.e., loads and rolling speed. The experimental results indicate that the proposed method outperforms other state-of-the-art methods in fault diagnosis under varying working conditions.</p> <p> </p>

SAR Ship–Iceberg Discrimination in Arctic Conditions Using Deep Learning

Maritime surveillance of the Arctic region is of growing importance as shipping, fishing and tourism are increasing due to the sea ice retreat caused by global warming. Ships that do not identify themselves with a transponder system, so-called dark ships, pose a security risk. They can be detected by SAR satellites, which can monitor the vast Arctic region through clouds, day and night, with the caveat that the abundant icebergs in the Arctic cause false alarms. We collect and analyze 200 Sentinel-1 horizontally polarized SAR scenes from areas with high maritime traffic and from the Arctic region with a high density of icebergs. Ships and icebergs are detected using a continuous wavelet transform, which is optimized by correlating ships to known AIS positions. Globally, we are able to assign 72% of the AIS signals to a SAR ship and 32% of the SAR ships to an AIS signal. The ships are used to construct an annotated dataset of more than 9000 ships and ten times as many icebergs. The dataset is used for training several convolutional neural networks, and we propose a new network which achieves state of the art performance compared to previous ship–iceberg discrimination networks, reaching 93% validation accuracy. Furthermore, we collect a smaller test dataset consisting of 424 ships from 100 Arctic scenes which are correlated to AIS positions. This dataset constitutes an operational Arctic test scenario. We find these ships harder to classify with a lower test accuracy of 83%, because some of the ships sail near icebergs and ice floes, which confuses the classification algorithms.

FlotGAIL: An operational adjustment framework for flotation circuits using generative adversarial imitation learning

In many industrial flotation processes, the operational adjustment is still manual according to individual experience. The adjustment of operational variables is mainly based on the discrimination of feed conditions and the observation of the appearance of flotation froth. Due to the limitation that partial properties of feed ore cannot be monitored online and the individual-level differences in operation experience, it is difficult to control the concentrate grade and recovery in the qualified range for a long time by manual operational adjustment. To resolve this problem, we propose FlotGAIL, an automatic operational adjustment framework based on generative adversarial imitation learning (GAIL) for the flotation circuits. From the perspective of an algorithm, this paper explores a novel way to learn operational adjustment policy. The epistemic uncertainty estimation of feed conditions provided by Bayesian neural networks is used as an additional guidance for the GAIL model to generate flotation operational trajectories, which it is just as an expert considers reducing the risk of process operation and increasing long-term rewards. In FlotGAIL, learning flotation operational adjustment from expert demonstrations is formulated as an imitation learning problem in a Markov decision process, and then the proposed model uses the signal provided by the discriminator as the reward function to guide the policy generator to complete the training process. From the perspective of engineering application, the experiments have verified that the proposed operational adjustment method has a better effect in controlling concentrate grade and recovery up to standard, and FlotGAIL can be used as an effective tool to provide decision support for operators of other similar industrial processes.

Pit organ-based infrared discrimination sensitivity and signal transduction in the Burmese python (Python molurus bivitattus)

Burmese pythons (Python molurus bivitattus) use a unique infrared (IR) targeting system to acquire prey, avoid predators and seek thermoregulatory sites through detection of IR energy in the environment. Previous studies of sensitivity of the python IR system that relied on analysis of complex, natural behaviors lacked robust, reliable responses in animals habituated to experiments, and in vitro electrophysiological study failed to test behavioral function of the implicated protein thermoreceptor, TRPA1. The present study used conditioned discrimination procedures to analyze behavioral sensitivity and signal transduction in the python IR system. Pythons trained to behaviorally discriminate thermal stimuli averaged 70% correct choices, but failed to make correct choices when pit organs were physically occluded with IR-blocking material. The pythons exhibited greater sensitivity to thermal stimuli than previously reported, evident by correct choices that exceeded chance in response to a 14 × 10−6 W cm−2 irradiance contrast, or 0.5 °C thermal differential. Finally, in a test of the behavioral role of the putative thermoreceptor protein TRPA1, despite pit organ treatment with a TRPA1 inhibitor, python performance exceeded chance and was similar to baseline discrimination and control trials. Collectively, the results suggest that the IR system is a high sensitivity, broad-spectrum thermosensor that may operate through different and/or multiple thermoreceptive proteins with overlapping spectral response profiles. The findings reported here provide a better understanding of the relationship between the brain, behavior and environment in driving survival and ecological success of the Burmese python, especially as an invasive megapredator in the southern United States.

Revisiting the use of two choices in the matching‐to‐sample procedure

The use of two choices in the matching-to-sample (MTS) procedure has been discouraged in the literature because it may lead to "reject control," resulting in failures to establish equivalence classes. In the present study, reject control was prevented during training with the two-choice MTS procedure by presenting the correct comparison with one of five possible incorrect comparisons across trials. This procedure was compared to a six-choice MTS procedure, in which these same six comparison stimuli were presented simultaneously across trials. In Experiment 1, conditional discrimination training and emergent relations testing maintained the same number of comparison choices, two or six. Experiment 2 assessed whether training with two or six choices would result in successful tests under a different configuration from the one with which training occurred (i.e., six or two choices, respectively). In Experiment 3, the conditions were the same as in Experiment 2, but minimal instructions were given to the participants. The results showed the establishment of equivalence classes in all test conditions, thus demonstrating success of the different training conditions. The two-choice MTS procedure appears to be at least as effective as the six-choice procedure for training conditional relations and establishing equivalence classes.

Periodic attention operates faster during more complex visual search

Attention has been found to sample visual information periodically, in a wide range of frequencies below 20 Hz. This periodicity may be supported by brain oscillations at corresponding frequencies. We propose that part of the discrepancy in periodic frequencies observed in the literature is due to differences in attentional demands, resulting from heterogeneity in tasks performed. To test this hypothesis, we used visual search and manipulated task complexity, i.e., target discriminability (high, medium, low) and number of distractors (set size), while electro-encephalography was simultaneously recorded. We replicated previous results showing that the phase of pre-stimulus low-frequency oscillations predicts search performance. Crucially, such effects were observed at increasing frequencies within the theta-alpha range (6–18 Hz) for decreasing target discriminability. In medium and low discriminability conditions, correct responses were further associated with higher post-stimulus phase-locking than incorrect ones, in increasing frequency and latency. Finally, the larger the set size, the later the post-stimulus effect peaked. Together, these results suggest that increased complexity (lower discriminability or larger set size) requires more attentional cycles to perform the task, partially explaining discrepancies between reports of attentional sampling. Low-frequency oscillations structure the temporal dynamics of neural activity and aid top-down, attentional control for efficient visual processing.

The First Baby Born After Polygenic Embryo Screening

The First Baby Born After Polygenic Embryo Screening

EHealth Interventions to Address HIV and Other Sexually Transmitted Infections, Sexual Risk Behavior, Substance Use, and Mental Ill-health in Men Who Have Sex With Men: Systematic Review and Meta-analysis

BackgroundMen who have sex with men experience disproportionately high levels of HIV and other sexually transmitted infections (STIs), sexual risk behavior, substance use, and mental ill-health. These experiences are interrelated, and these interrelations are potentiated by structural conditions of discrimination, stigma, and unequal access to appropriate health services, and they magnify each other and have intersecting causal pathways, worsening both risk for each condition and risk for the negative sequelae of each condition. eHealth interventions could address these issues simultaneously and thus have wide-ranging and greater effects than would be for any 1 outcome alone.ObjectiveWe systematically reviewed the evidence for the effectiveness of eHealth interventions in addressing these outcomes separately or together.MethodsWe searched 19 databases for randomized trials of interactive or noninteractive eHealth interventions delivered via mobile phone apps, internet, or other electronic media to populations consisting entirely or principally of men who have sex with men to prevent HIV, STIs, sexual risk behavior, alcohol and drug use, or common mental illnesses. We extracted data and appraised each study, estimated meta-analyses where possible by using random effects and robust variance estimation, and assessed the certainty of our findings (closeness of the estimated effect to the true effect) by using GRADE (Grading of Recommendations, Assessment, Development and Evaluations).ResultsWe included 14 trials, of which 13 included active versus control comparisons; none reported mental health outcomes, and all drew from 12 months or less of follow-up postintervention. Findings for STIs drew on low numbers of studies and did not suggest consistent short-term (<3 months postintervention; d=0.17, 95% CI –0.18 to 0.52; I2=0%; 2 studies) or midterm (3-12 months postintervention, no meta-analysis, 1 study) evidence of effectiveness. Eight studies considering sexual risk behavior outcomes suggested a short-term, nonsignificant reduction (d=–0.14, 95% CI –0.30 to 0.03) with very low certainty, but 6 studies reporting midterm follow-ups suggested a significant impact on reducing sexual risk behavior (d=–0.12, 95% CI –0.19 to –0.05) with low certainty. Meta-analyses could not be undertaken for alcohol and drug use (2 heterogeneous studies) or for HIV infections (1 study for each of short-term or midterm follow-up), and alcohol outcomes alone were not captured in the included studies. Certainty was graded as low to very low for most outcomes, including all meta-analyses.ConclusionsTo create a comprehensive eHealth intervention that targets multiple outcomes, intervention evaluations should seek to generalize both mechanisms and components that are successfully used to achieve change in 1 outcome over multiple outcomes. However, additional evaluations of interventions seeking to address outcomes other than sexual risk behavior are needed before development and evaluation of a joined-up intervention.