Category Cohesion And Distinctiveness Favor Different .
Category Cohesion and DistinctivenessFavor Different Taxonomic LevelsAcross Human Visual CortexMarius Cătălin IordanMichelle R. GreeneDiane M. BeckAttention andPerception LabSfN 2014Li Fei-Fei
Taxonomic Levels in Human Visual Cortex2Iordan, Greene, Beck, Fei-Fei
How do category representations change acrosstaxonomic levels in human visual cortex ?GeneralSUPERORDINATEnatural objectBASICdogSUBORDINATEborder collieSpecificTaxonomic Levels in Human Visual Cortex2Rosch et al. (1976)Iordan, Greene, Beck, Fei-Fei
How do category representations change acrosstaxonomic levels in human visual cortex ?1.Experimental Setup2.Similarity of NeuralPatterns3.Category Information atEach Taxonomic LevelGeneralSUPERORDINATEnatural objectBASICdogSUBORDINATEborder collieSpecificIs there a privileged taxonomic level ?Taxonomic Levels in Human Visual Cortex3Iordan, Greene, Beck, Fei-Fei
big, inanimate objectssubset of Rosch et al. (1976)UTRSINNTMETaxonomic Levels in Human Visual CortexFURNITUREVEHICLE4Iordan, Greene, Beck, Fei-Fei
big, inanimate objectssubset of Rosch et al. (1976)UTRSINNTMETaxonomic Levels in Human Visual CortexFURNITUREVEHICLE5Iordan, Greene, Beck, Fei-Fei
40VEHICLENTMEFURNITUREUTRSIN4040Taxonomic Levels in Human Visual Cortex6Iordan, Greene, Beck, Fei-Fei
Match-to-Category Behavioral Experimentis this a “fighter plane” / “plane” / “vehicle” ?Taxonomic Levels in Human Visual Cortex7Iordan, Greene, Beck, Fei-Fei
Match-to-Category Behavioral ExperimentTaxonomic Levels in Human Visual Cortex8Iordan, Greene, Beck, Fei-Fei
fMRI Experiment: Example “Fighter Plane” Block8 images per block 5 blocks per category 27 categories Taxonomic Levels in Human Visual Cortex9Iordan, Greene, Beck, Fei-Fei
27 categoriesx40 images eachVEHICLENTMEFURNITUREUTRSINpassive viewingfMRIexperiment3taxonomiclevelsTaxonomic Levels in Human Visual Cortex10Iordan, Greene, Beck, Fei-Fei
27 categoriesx40 images eachSUBORDINATEVEHICLENTMEFURNITUREUTRSINpassive viewingfMRIexperiment3taxonomiclevelsTaxonomic Levels in Human Visual Cortex10Iordan, Greene, Beck, Fei-Fei
27 categoriesx40 images eachSUBORDINATEVEHICLEpassive ATE3taxonomiclevelsTaxonomic Levels in Human Visual Cortex10Connolly et al. (2012), Konkle & Caramazza (2013)Iordan, Greene, Beck, Fei-Fei
27 categoriesx40 images eachSUBORDINATEBASICVEHICLEpassive ATE3taxonomiclevelsTaxonomic Levels in Human Visual Cortex10Rosch et al. (1976)Connolly et al. (2012), Konkle & Caramazza (2013)Iordan, Greene, Beck, Fei-Fei
How category representations change acrosstaxonomic levels in human visual cortex1. Experimental Setup stimulus sets & fMRI scanning2. Similarity of Neural Patterns within- and between-category similarity3. Category Information at Each Taxonomic Level MVPA decodingTaxonomic Levels in Human Visual Cortex11Iordan, Greene, Beck, Fei-Fei
Category Boundary Effect: Basic Level ExampleWithin-Category SimilarityACohesionCategory Boundary Effect mean( A )mean(B )computed separatelyfor eachtaxonomic levelBetween-Category SimilarityBDistinctivenessTaxonomic Levels in Human Visual Cortex12Kriegeskorte et al. (2008)Iordan, Greene, Beck, Fei-Fei
Category Boundary Effect: ResultsSubordinateCATEGORY BOUNDARY EFFECT0.06*** ***** ***n.s.SuperordinateBasicn.s. n.s.n.s. **** *******0.030.03********************0.02********** ***0.02********0.010.00n 17* p 0.05** p 0.01*** p 0.0010.060.010.00V1V2V3vEarly VisualTaxonomic Levels in Human Visual CortexhV4LOCObjects13PPATOSRSCScenesIordan, Greene, Beck, Fei-Fei
Gradualtrade-offin favorEffect:of theResultsbasic levelCategoryBoundarySubordinateCATEGORY BOUNDARY EFFECT0.06*** ***** ***n.s.SuperordinateBasicn.s. n.s.n.s. **** *******0.030.03********************0.02********** ***0.02********0.010.00n 17* p 0.05** p 0.01*** p 0.0010.060.010.00V1V2V3vEarly VisualTaxonomic Levels in Human Visual CortexhV4LOCObjects14PPATOSRSCScenesIordan, Greene, Beck, Fei-Fei
Gradualtrade-offin favorEffect:of theResultsbasic levelCategoryBoundarySubordinateCATEGORY BOUNDARY EFFECT0.06*** ***** ***n.s.SuperordinateBasicn.s. n.s.n.s. **** *******0.030.03********************0.02********** ***0.02********0.010.00n 17* p 0.05** p 0.01*** p 0.0010.060.010.00V1V2V3vEarly VisualTaxonomic Levels in Human Visual CortexhV4LOCObjects14PPATOSRSCScenesIordan, Greene, Beck, Fei-Fei
Gradual trade-off in favor of the basic levelCategory Boundary Effect DifferenceBasic Level Subordinate Level0.06p 0.0001, Friedman ic Levels in Human Visual CortexV3v15hV4LOCIordan, Greene, Beck, Fei-Fei
Gradual trade-off in favor of the basic levelCategory Boundary Effect DifferenceBasic Level Superordinate Level0.03***p 0.0001, Friedman test0.020.010.00-0.01-0.02***V1V2-0.03Taxonomic Levels in Human Visual CortexV3v16hV4LOCIordan, Greene, Beck, Fei-Fei
How category representations change acrosstaxonomic levels in human visual cortex1. Experimental Setup stimulus sets & fMRI scanning2. Similarity of Neural Patterns within- and between-category similarity3. Category Information at Each Taxonomic Level MVPA decodingTaxonomic Levels in Human Visual Cortex17Iordan, Greene, Beck, Fei-Fei
MVPA Decoding: Resultsdecode category at each taxonomic level independentlySubordinatePROPORTION CORRECTABOVE CHANCEns*ns** *ns ns0.12* *Superordinate*ns*ns* Chancen 17* p 0.05** p 0.010.03ChanceV1V2V3vEarly VisualhV4LOCObjectsPPATOSRSCScenes*** p 0.001Taxonomic Levels in Human Visual Cortex18Iordan, Greene, Beck, Fei-Fei
Basic level is theoptimallevel ofspecificity in LOCMVPADecoding:Resultsdecode category at each taxonomic level independentlySubordinatePROPORTION CORRECTABOVE CHANCEns*ns** *ns ns0.12* *Superordinate*ns*ns* Chancen 17* p 0.05** p 0.010.03ChanceV1V2V3vEarly VisualhV4LOCObjectsPPATOSRSCScenes*** p 0.001Taxonomic Levels in Human Visual Cortex18Iordan, Greene, Beck, Fei-Fei
How category representations change acrosstaxonomic levels in visual cortexreal-world object taxonomywith behavioral basic level advantageINSUTRNMETFURNITUREVEHICLEgradual trade-off between basic and other levelsin favor of the formerCategory Boundary Effect DifferenceBasic Level Superordinate Level0.03***p asic level is optimal level of specificitydecodable from LOC patternsTaxonomic Levels in Human Visual Cortex19Iordan, Greene, Beck, Fei-Fei
How category representations change acrosstaxonomic levels in visual cortexreal-worldobject taxonomybasic levelrepresentationmay be anwith behaviora
FURNITURE VEHICLE T Taxonomic Levels in Human Visual Cortex 20 Iordan, Greene, Beck, Fei-Fei real-world object taxonomy with behavioral basic level advantage gradual trade-off between subordinate and basic levels in favor of the latter basic level is optimal level of specificity decodab
Territorial Cohesion (CEC 2008), it makes no reference to the debate on territorial cohesion however. Like Barca, that debate conceives of territory as a platform for co-ordinating public and private initiatives. Much as Durkheim (English edition 1933) conceives of social cohesion as an element of the social division of labour, so too is the
Varying Sentence Openers for Emphasis, Pace, and Cohesion, Spring 2015. 3 of 5 Emphasis, Pace, and Cohesion Now that we have looked at different ways to start a sentence, let’s examine how they affect emphasis, pace, and cohesion. Emphasis The end of the main clause is the most emphatic position of a
Keywords: territorial, EU cohesion, CAP, Pillar 1, Pillar 2, rural development policy. 1 Introduction The Second Report on Economic and Social Cohesion (European Commission, 2001) called for cohesion policy to promote a more balanced and more sustainable development of the European territory, in line with the European Spatial
of territorial cohesion is not explicitly used in the text, it is implicit via references made to the ‘European Regions’. However, a probable reason not to include the notion of territorial cohesion at this stage of EU poli-tics is the fact that territorial cohesion policy is a sensitive issue for the sovereignty of states.
Category 2: Goals for coaching session - 5 questions Category 9: Awareness on obstacles - 7q Category 3: Clarifying needs - 3 questions Category 10: facing fears - 12 questions Category 4: Listening/Focus - 26 questions Category 11: Client stuck/ confusion -29 q Category 5: Options/ Choices - 3
The favor of God will change your circumstances! The favor of God will open doors for you, and it will even change rules, regulations, and policies for your benefit. One encounter with the favor of God is worth a lifetime of labor! It's important for you to understand that although the favor of God is available for us to walk in on a .
Favor of the Faith Preliminary Investigation (Revised May 2016) This information is intended to guide those who may wish to petition for the use of the favor of the faith. The dissolution of a non-sacramental marriage by the Pope as the Vicar of Christ can be done, according to the norm of law, in order to favor the faith of a Catholic party.
The Automotive Sector Deal, the first in a rolling series of intended deals with the sector, builds on the partnership between the government and industry that has been in place since the Automotive Council was established in 2009, setting the direction and long-term strategic priorities for the sector. This partnership has yielded results: vehicle and engine output has increased, productivity .
