Search

BackgroundChoroid plexus (ChP) enlargement has been described in first-episode psychosis and psychosis spectrum disorders, but whether ChP enlargement occurs before disease onset is unknown. This study investigated whether ChP volume is enlarged in individuals with clinical high-risk (CHR) for psychosis and whether these changes are related to clinical, cortical, and plasma analyte measures.MethodsClinical and neuroimaging data from the North American Prodrome Longitudinal Study 2 (NAPLS2) was used for analysis. A total of 509 participants (169 controls, 340 CHR) were recruited across 8 sites. Conversion status was determined until 2-years of follow-up, with 36 patients developing psychosis. The lateral ventricle ChP was manually segmented from all available baseline brain scans. A subsample of 84 participants (31 controls, 53 CHR) had plasma analyte and neuroimaging data.ResultsCompared to controls, CHR overall (d=0.22, p=0.017) and converters(d=0.21, p=0.041)demonstrated higher ChP volumes, but not nonconverters. In CHR, greater ChP volume correlated with lower cortical(r=−0.22, pand subcortical gray matter volume(r=− 0.21, p, total white matter volume(r=−0.28,p, and larger lateral ventricle volume(r=0.63,p. In CHR, greater ChP volume, but not lateral ventricle volume, correlated with higher C3(r=0.39, p=0.005)and TSH(r=0.31, p=0.026), and lower MMP7(r=−0.30, p=0.032)and UMOD(r=−0.33, p=0.019).Conclusions and RelevanceCHR and converters to psychosis demonstrated significantly larger ChP volumes compared to controls. ChP enlargement was associated with cortical volume reduction and increased peripheral inflammatory markers. These findings suggest that the ChP may be a key biomarker in psychosis disease onset and conversion.

BackgroundCannabis use (CA) and childhood trauma (CT) independently increase the risk of earlier psychosis onset; but their interaction in relation to psychosis risk and association with endocannabinoid-receptor rich brain regions, i.e. the hippocampus (HP), remains unclear. The objective was to determine whether lower age of psychosis onset (AgePsyOnset) is associated with CA and CT through mediation by the HP, and genetic risk, as measured by schizophrenia polygene scores (SZ-PGRS).MethodsCross- sectional, case-control, multicenter sample from 5 metropolitan US regions. Participants (n=1185) included 397 controls not affected by psychosis (HC); 209 participants with bipolar disorder type-1; 279 with schizoaffective disorder; and 300 with schizophrenia (DSM IV-TR). CT was assessed using the Childhood Trauma Questionnaire (CTQ); CA was assessed by self-reports and trained clinical interviewers. Assessment included neuroimaging, symptomatology, cognition and calculation of the SZ polygenic risk score (SZ-PGRS).OutcomesIn survival analysis, low CT and CA are associated with lower AgePsyOnset. At high CT or CA, CT or CA are individually sufficient to affect AgePsyOnset. CT relation with AgePsyOnset is mediated in part by the HP in CA users before AgePsyOnset. CA before AgePsyOnset is associated with higher SZ-PGRS and correlated with younger age at CA usage.InterpretationCA and CT interact to increase risk when moderate; while severe CT and/or CA abuse/dependence are each sufficient to affect AgePsyOnset, indicating a ceiling effect. Probands with/out CA before AgePsyOnset differ on biological variables, suggesting divergent pathways to psychosis.FundingMH077945; MH096942; MH096913; MH077862; MH103368; MH096900; MH122759.Research in contextEvidence before this studyCannabis use (CA) and childhood trauma (CT) independently increase the risk of earlier development of psychosis. Scarce evidence exists on the interaction between CA and CT and the neurobiological substrate of their interaction.Added value of this studyAnalysis of a large transdiagnostic sample of psychosis probands and controls (N=1288) indicates synergy of CT and CA and small but significant contribution of the posterior hippocampus. Data further indicate existence of two populations of probands with psychosis, those with and those without CA after CT before psychosis onset. CT and CA before psychosis onset interact according to a stepwise increase up to reaching a ceiling effect.Implications of all the available evidenceClinically, youth with low, medium CT need to be targeted for intervention before CA onset.

BACKGROUND Neurovascular abnormalities are relevant to the pathophysiology of bipolar disorder (BD), which can be assessed using cerebral blood flow (CBF) imaging. CBF alterations have been identified in BD, but studies to date have been small and inconclusive. We aimed to determine cortical gray matter CBF (GM-CBF) differences between BD and healthy controls (HC) and to identify relationships between CBF and clinical or cognitive measures. METHODS Cortical GM-CBF maps were generated using Pseudo-Continuous Arterial Spin Labeling (pCASL) for 109 participants (BD, n = 61; HC, n = 48). We used SnPM13 to perform non-parametric voxel-wise two-sample t-tests comparing CBF between groups. We performed multiple linear regression to relate GM-CBF with clinical and cognitive measures. Analysis was adjusted for multiple comparisons with 10,000 permutations. Significance was set at a voxel level threshold of P

BACKGROUND: Peripheral inflammation is implicated in schizophrenia, however, not all individuals demonstrate inflammatory alterations. Recent studies identified inflammatory subtypes in chronic psychosis with high inflammation having worse cognitive performance and displaying neuroanatomical enlargement compared to low inflammation subtypes. It is unclear if inflammatory subtypes exist earlier in the disease course, thus, we aim to identify inflammatory subtypes in antipsychotic naïve First-Episode Schizophrenia (FES). METHODS: 12 peripheral inflammatory markers, clinical, cognitive, and neuroanatomical measures were collected from a naturalistic study of antipsychotic-naïve FES patients. A combination of unsupervised principal component analysis and hierarchical clustering was used to categorize inflammatory subtypes from their cytokine data (17 FES High, 30 FES Low, and 33 healthy controls (HCs)). Linear regression analysis was used to assess subtype differences. Neuroanatomical correlations with clinical and cognitive measures were performed using partial Spearman correlations. Graph theoretical analyses were performed to assess global and local network properties across inflammatory subtypes. RESULTS: The FES High group made up 36% of the FES group and demonstrated significantly greater levels of IL1β, IL6, IL8, and TNFα compared to FES Low, and higher levels of IL1β and IL8 compared to HCs. FES High had greater right parahippocampal, caudal anterior cingulate, and bank superior sulcus thicknesses compared to FES Low. Compared to HCs, FES Low showed smaller bilateral amygdala volumes and cortical thickness. FES High and FES Low groups demonstrated less efficient topological organization compared to HCs. Individual cytokines and/or inflammatory signatures were positively associated with cognition and symptom measures. CONCLUSIONS: Inflammatory subtypes are present in antipsychotic-naïve FES and are associated with inflammation-mediated cortical expansion. These findings support our previous findings in chronic psychosis and point towards a connection between inflammation and blood-brain barrier disruption. Thus, identifying inflammatory subtypes may provide a novel therapeutic avenue for biomarker-guided treatment involving anti-inflammatory medications.

Impairments of the visual system are implicated in psychotic disorders. However, studies exploring visual cortex (VC) morphology in this population are limited. Using data from the Bipolar-Schizophrenia Network on Intermediate Phenotypes consortium, we examined VC structure in psychosis probands and their first-degree relatives (RELs), sex differences in VC measures, and their relationships with cognitive and peripheral inflammatory markers.Cortical thickness, surface area, and volume of the primary (Brodmann area 17/V1) and secondary (Brodmann area 18/V2) visual areas and the middle temporal (V5/MT) region were quantified using FreeSurfer version 6.0 in psychosis probands (n = 530), first-degree RELs (n = 544), and healthy control subjects (n = 323). Familiality estimates were determined for probands and RELs. General cognition, response inhibition, and emotion recognition functions were assessed. Systemic inflammation was measured in a subset of participants.Psychosis probands demonstrated significant area, thickness, and volume reductions in V1, V2, and MT, and their first-degree RELs demonstrated area and volume reductions in MT compared with control subjects. There was a higher degree of familiality for VC area than thickness. Area and volume reductions in V1 and V2 were sex dependent, affecting only female probands in a regionally specific manner. Reductions in some VC regions were correlated with poor general cognition, worse response inhibition, and increased C-reactive protein levels.The visual cortex is a site of significant pathology in psychotic disorders, with distinct patterns of area and thickness changes, sex-specific and regional effects, potential contributions to cognitive impairments, and association with C-reactive protein levels.

Schizophrenia, Schizoaffective, and Bipolar disorders share behavioral and phenomenological traits, intermediate phenotypes, and some associated genetic loci with pleiotropic effects. Volumetric abnormalities in brain structures are among the intermediate phenotypes consistently reported associated with these disorders. In order to examine the genetic underpinnings of these structural brain modifications, we performed genome-wide association analyses (GWAS) on 60 quantitative structural brain MRI phenotypes in a sample of 777 subjects (483 cases and 294 controls pooled together). Genotyping was performed with the Illumina PsychChip microarray, followed by imputation to the 1000 genomes multiethnic reference panel. Enlargement of the Temporal Horns of Lateral Ventricles (THLV) is associated with an intronic SNP of the gene NRXN1 (rs12467877, P = 6.76E–10), which accounts for 4.5% of the variance in size. Enlarged THLV is associated with psychosis in this sample, and with reduction of the hippocampus and enlargement of the choroid plexus and caudate. Eight other suggestively significant associations (P

Theory of Mind (ToM) refers to the ability to perceive others' mental states. Lower ToM has often been associated with poorer functional outcomes in schizophrenia, making it an important treatment target. However, little is known about the underlying neural mechanisms associated with ToM impairments in early course schizophrenia. This study aimed to validate the False Belief task to measure ToM in schizophrenia and to identify aberrant brain activity associated with impairments. 36 individuals with early course schizophrenia and 17 controls were administered the Hinting Task and performed a functional magnetic resonance imaging (fMRI) False Belief task. Between-group differences were examined in a priori regions of interest (ROIs) known to be associated with ToM tasks: medial prefrontal cortex, ventral medial prefrontal cortex, and both the left and right temporal parietal junction (TPJ). We observed a significant positive association between Hinting Task performance and False Belief accuracy, validating the False Belief task as a measure of ToM. Compared to controls, individuals with schizophrenia exhibited reduced brain activation in all four ROIs during the fMRI False Belief task. Furthermore, task-related activations in bilateral TPJs were shown to be positively associated with ToM abilities regardless of diagnosis. Individuals with schizophrenia with lower performance on the False Belief task showed significant reductions in task-related activation in the bilateral TPJ compared to controls, while reductions were not significant for those with higher performance. Our findings suggest that lower neural activity in the bilateral TPJ are associated with ToM impairments observed in individuals with early course schizophrenia.

The thalamus, amygdala, and hippocampus play important pathophysiologic roles in psychosis. Few studies have prospectively examined subcortical nuclei in relation to predicting clinical outcomes after a first-episode of psychosis (FEP). Here, we examined volumetric differences and trajectories among subcortical nuclei in FEP patients and their associations with illness severity. Clinical and brain volume measures were collected using a 1.5T MRI scanner and processed using FreeSurfer 6.0 from a prospective study of antipsychotic-naive FEP patients of FEP-schizophrenia (FEP-SZ) (baseline, n = 38; follow-up, n = 17), FEP non-schizophrenia (FEP-NSZ) (baseline, n = 23; follow-up, n = 13), and healthy controls (HCs) (baseline, n = 47; follow-up, n = 29). Compared to FEP-NSZ and HCs, FEP-SZ had significantly smaller thalamic anterior nuclei volume at baseline. Longitudinally, FEP-SZ showed a positive rate of change in the amygdala compared to controls or FEP-NSZ, as well as in the basal, central and accessory basal nuclei compared to FEP-NSZ. Enlargement in the thalamic anterior nuclei predicted a worsening in overall psychosis symptoms. Baseline thalamic anterior nuclei alterations further specify key subcortical regions associated with FEP-SZ pathophysiology. Longitudinally, anterior nuclei volume enlargement may signal symptomatic worsening. The amygdala and thalamus structures may show diagnostic differences between schizophrenia and non-schizophrenia psychoses, while the thalamus changes may reflect disease or treatment related changes in clinical outcome.

Phishing is a challenging problem that has been addressed by many researchers in several papers using many different datatsets and techniques~\citedas2019sok. Researchers usually test their proposed methods with limited metrics, datasets, and parameters when presenting new features or approach(es). Hence, the need arises for a benchmarking framework and dataset to evaluate such systems as comprehensively as possible. In this paper, we discuss: (i) our efforts on the creation and dissemination of diverse and representative datasets for phishing email, website and URL detection, and (ii) PhishBench, our framework for benchmarking phishing detection systems. PhishBench allows researchers to evaluate and compare features and classification approaches easily and efficiently on the provided data.

Background The hippocampus (HP) is affected across psychoses, including schizophrenia (SZ), bipolar type 1 (BDP) and schizoaffective (SAD) disorders. We examined HP subfield volumetric abnormalities along the anterior-posterior (ventral-dorsal) axis of the HP in psychosis probands, defined by traditional (DSM) diagnoses and biologically defined subtypes (biotypes, based on cognition and electrophysiology). We hypothesized that biotypes would be better discriminated by HP longitudinal axis subfields abnormalities than DSM. Methods The sample included 455 probands from the Bipolar Schizophrenia Network for intermediate Phenotypes (BSNIP) dataset (age 35 ± 12.0): 124 unaffected (age 40.4 ± 15.8) and 299 healthy controls (HC; 37 ± 12.0). Probands were: SZ (190), BDP (151), SAD (114). Probands according to B-SNIP defined biotypes were: biotype-1 (BT1; 120), biotype-2 (BT2; 145), biotype-3 (BT3, 190). 3 T MRI scans were processed with FreeSurfer6.0. The anterior (aHP) and posterior (pHP) HP and aHP and pHP subfields were extracted. Cognitive and clinical data were collected. Results All biotypes had smaller aHP subfields compared to HC. BT1 had smaller aHP than both BT2 and BT3. pHP subfields were also smaller in BT1 compared to HC and BT3, while the granule cells layer of the dentate gyrus distinguished BT2 from HC. DSM: aHP subfields were smaller in all DSM types compared to HC and did not differ among DSM categories. A few pHP subfields were affected in SAD and SZ compared to HC and distinguished SAD and SZ from BDP. Probands had smaller aHP compared to unaffected relatives. Conclusions Differences in subfield volumetric abnormalities along the anterior- posterior axis of the HP exist across psychoses. aHP abnormalities differ between psychosis probands and HC but do not discriminate among DSM categories. In contrast, biotypes can be differentiated from HC and from each other according to aHP-pHP subfields volumetric abnormalities. Thus, biotype typology may better reflect underlying neurobiology.

Background Symptoms of mental diseases can be cross-diagnostic. For example psychosis is not limited to schizophrenia (SZ), but can be also present in bipolar (BPP) and schizoaffective (SA) disorders. Morphometric findings are often also cross-diagnostic, For example, volume decreases of the hippocampus (HP), essential to memory functions, is affected in both schizophrenia and bipolar disorder. Biotyping, grouping of patients according to biomarker-based categories is one attempt to approach cross-diagnostic features in the hope of better classification and ultimately tailored interventions. Here we combine biotyping, in addition to traditional diagnosing, with analysis of the HP volume. Furthermore, we explore the HP according to its anterior (aHP) versus posterior (pHP) anatomical division, following the functional division of this medial temporal lobe structure. Our hypothesis is that aHP and pHP are differentially affected in biotypes and diagnostic groups and reflective of differences in cognitive functions affected in each biotype and/or diagnostic group. Methods The sample included 450 probands (age 35±12.0 yo) and 325 healthy controls (HC; 37±12.0 yo). Of the probands 187 were diagnosed with SZ, 150 with BPP (psychosis type), 113 with SA disorders; after biotyping the same probands were categorized into three biotypes: 118 biotype 1 (BP1), 142 biotype 2 (BP2), 187 biotype 3 (BP3). 3T MRI data were quality checked and processed with FreeSurfer 6.0. The HP measures extracted were: the whole HP, and further HP subfields according to aHP and pHP. aHP and pHP subfields were reconstituted into left and right aHP and pHP. BACS cognitive data and PANSS, SAPS, MADRS and mania clinical ratings were collected. Results Analysis According to Diagnoses: There were no aHP or pHP volume differences among diagnostic groups, but all diagnostic groups demonstrated volume decreases of the left and right aHP compared to HC with effect sizes (Cohen’s d) between 0.3 and 0.6. Further, SA had decreases of the left and right pHP as well and SZ of the left pHP (effect sizes 0.3–0.4) compared to HC. BPP pHP volume were not different than pHP volumes measured in HC. Analysis According to Biotyping: Decreases of left and right aHP and pHP volumes were present in BP1 compared to HC while decreases of left pHP and left and right aHP were present in BT2 compared to HC. Only aHP was affected in BT3 compared to HC. Left and right aHP and pHP distinguished BT1 and BT3. BT1 had further decreased left pHP volume compared to BP2. Several anterior portions of the subfields volumes were driving the aHP vs pHP differences among biotypes while differences of the posterior portions of the subfields were fewer. aHP differences between BP1 and BP2 were driven by the anterior portion of the right subiculum. Discussion Biotypes are more effective than diagnoses in discriminating abnormalities of the HP across psychosis categories. Further, analysis of the aHP versus the pHP demonstrates that the HP should be analyzed not as a whole, or as its subfields, but according to anterior versus posterior divisions of its subfields and to the sum of anterior versus posterior subfields in aHP and pHP. Further studies need to elucidate the cognitive functional correlates of aHP- pHP differences across psychoses and healthy populations.

Visual perceptual and processing deficits are common in schizophrenia and possibly point towards visual pathway alterations. However, no studies have examined visual cortical morphology in first-episode psychosis (FEP). In an antipsychotic-naive FEP population, we investigated primary visual (V1), association area (V2), and motion perception (V5/MT) morphology compared to controls. We found reductions in the V1 and V2 areas, greater MT area and lower MT thickness in the FEP-schizophrenia group when compared to controls. Also, lower MT thickness was associated with worse negative symptoms. Our results shed light on this poorly studied area of visual cortex morphology in FEP.

Techniques from data science are increasingly being applied by researchers to security challenges. However, challenges unique to the security domain necessitate painstaking care for the models to be valid and robust. In this paper, we explain key dimensions of data quality relevant for security, illustrate them with several popular datasets for phishing, intrusion detection and malware, indicate operational methods for assuring data quality and seek to inspire the audience to generate high quality datasets for security challenges.

Subtypes of schizophrenia, constructed using clinical phenomenology to resolve illness heterogeneity, have faced criticism due to overlapping symptomatology and longitudinal instability; they were therefore dropped from the Diagnostic Statistical Manual-5. Cognitive and imaging findings comparing paranoid (P-SZ) and non-paranoid (disorganized, residual and undifferentiated; NP-SZ) schizophrenia have been limited due to small sample sizes. We assessed P-SZ and NP-SZ using symptomatology, cognition and brain structure and predicted that there would be few neurobiological differences. P-SZ (n = 237), NP-SZ (n = 127) and controls (n = 430) were included from a multi-site study. In a subset of this sample, structural imaging measures (P-SZ, n = 133; NP-SZ, n = 67; controls, n = 310) were calculated using Freesurfer 6.0. Group contrasts were run using analysis of covariance, controlling for age, sex, race and site, p-values were corrected using False Discovery Rate (FDR) and were repeated excluding the residual subtype. Compared to NP-SZ (with and without the residual subtype), P-SZ displayed fewer negative symptoms, faster speed of processing, larger bilateral hippocampus, right amygdala and their subfield volumes. Additionally, NP-SZ (with residual subtype) displayed fewer depressive symptoms and higher left transverse temporal cortical thickness (CT) but NP-SZ without residual subtype showed lower GAF scores and worse digit sequencing compared to P-SZ. No differences in positive symptoms and functioning (global or social) were detected. Subtle but significant differences were seen in cognition, symptoms, CT and subcortical volumes between P-SZ and NP-SZ. While the magnitude of these differences is not large enough to justify them as distinct categories, the paranoid- nonparanoid distinction in schizophrenia merits further investigation.

Studies utilizing optical coherence tomography (OCT) in psychosis have identified abnormalities in retinal cytoarchitecture. We aim to analyze retinal layer topography in psychosis and its correlation with clinical and imaging parameters. Macular retinal images were obtained via OCT in psychosis probands (n = 25) and healthy controls (HC, n = 15). Clinical, cognitive and structural MRI data were collected from participants. No thinning was noted for the retinal nerve fiber, ganglion cell or inner plexiform layers. We found significant thinning in the right inner temporal, right central, and left inner superior quadrants of the outer nuclear layer (ONL) in probands compared to HC. Thickening of the outer plexiform layer (OPL) was observed in the right inner temporal, left inner superior, and left inner temporal quadrants. The right inner temporal and left inner superior quadrants of both the OPL and ONL showed significant inverse correlations. Retinal pigment epithelium thinning correlated with worse mania symptoms, and thinning in the ONL was associated with worse cognitive function. ONL thinning was also associated with smaller total brain and white matter volume. Our findings suggest that outer retinal layers may provide additional insights into the pathophysiology of psychosis, possibly reflecting synaptic or inflammatory aberrations that lead to retinal pathologies.

The increased throughput and decreased cost of next-generation sequencing (NGS) have shifted the bottleneck genomic research from sequencing to annotation, analysis and accessibility. This is particularly challenging for research communities working on organisms that lack the basic infrastructure of a sequenced genome, or an efficient way to utilize whatever sequence data may be available. Here we present a new database, the Assembled Searchable Giant Arthropod Read Database (ASGARD). This database is a repository and search engine for transcriptomic data from arthropods that are of high interest to multiple research communities but currently lack sequenced genomes. We demonstrate the functionality and utility of ASGARD using de novo assembled transcriptomes from the milkweed bug Oncopeltus fasciatus, the cricket Gryllus bimaculatus and the amphipod crustacean Parhyale hawaiensis. We have annotated these transcriptomes to assign putative orthology, coding region determination, protein domain identification and Gene Ontology (GO) term annotation to all possible assembly products. ASGARD allows users to search all assemblies by orthology annotation, GO term annotation or Basic Local Alignment Search Tool. User-friendly features of ASGARD include search term auto-completion suggestions based on database content, the ability to download assembly product sequences in FASTA format, direct links to NCBI data for predicted orthologs and graphical representation of the location of protein domains and matches to similar sequences from the NCBI non-redundant database. ASGARD will be a useful repository for transcriptome data from future NGS studies on these and other emerging model arthropods, regardless of sequencing platform, assembly or annotation status. This database thus provides easy, one-stop access to multi-species annotated transcriptome information. We anticipate that this database will be useful for members of multiple research communities, including developmental biology, physiology, evolutionary biology, ecology, comparative genomics and phylogenomics. Database URL: asgard.rc.fas.harvard.edu.

Background Arthropods are the most diverse animal phylum, but their genomic resources are relatively few. While the genome of the branchiopod Daphnia pulex is now available, no other large-scale crustacean genomic resources are available for comparison. In particular, genomic resources are lacking for the most tractable laboratory model of crustacean development, the amphipod Parhyale hawaiensis. Insight into shared and divergent characters of crustacean genomes will facilitate interpretation of future developmental, biomedical, and ecological research using crustacean models. Results To generate a transcriptome enriched for maternally provided and zygotically transcribed developmental genes, we created cDNA from ovaries and embryos of P. hawaiensis. Using 454 pyrosequencing, we sequenced over 1.1 billion bases of this cDNA, and assembled them de novo to create, to our knowledge, the second largest crustacean genomic resource to date. We found an unusually high proportion of C2H2 zinc finger-containing transcripts, as has also been reported for the genome of the pea aphid Acyrthosiphon pisum. Consistent with previous reports, we detected trans-spliced transcripts, but found that they did not noticeably impact transcriptome assembly. Our assembly products yielded 19,067 unique BLAST hits against nr (E-value cutoff e-10). These included over 400 predicted transcripts with significant similarity to D. pulex sequences but not to sequences of any other animal. Annotation of several hundred genes revealed P. hawaiensis homologues of genes involved in development, gametogenesis, and a majority of the members of six major conserved metazoan signaling pathways. Conclusions The amphipod P. hawaiensis has higher transcript complexity than known insect transcriptomes, and trans-splicing does not appear to be a major contributor to this complexity. We discuss the importance of a reliable comparative genomic framework within which to consider findings from new crustacean models such as D. pulex and P. hawaiensis, as well as the need for development of further substantial crustacean genomic resources.

Sports-related event sponsoring has rapidly grown into a key ingredient of the marketing communications mix. We use event study methodology to assess the net economic value of 2008 Beijing Olympic Games sponsorships. We find that investors judge the benefits that accrue to sponsoring companies to be commensurate with the expenses, as evidenced by insignificant announcement date abnormal returns. Furthermore, on the Games opening ceremony date the domestic sponsors’ share prices fall significantly while the international sponsors on average experience positive returns. The domestic firms’ sponsorship decision may have been based on national pride and emotional commitment, rather than on profit maximization.