Week 8 & 9 - ROC ROC ROC & Forgetting

• When constructing a fair lineup, the fillers should adhere to who’s description?

• Does every candidate on the lineup have to match the description perfectly?

• The fillers should match the physical description provided by the eyewitness, so the suspect does not stand out

• Details that are not mentioned by the witness are allowed to exhibit variation in the lineup

what is an example of an unfair lineup?

• the type of lineup where the suspect stands out from the rest.

  • example shown in class was of a black man on a lineup of white men and women

what is the diagnosticity ratio failing to capture?

• Did not capture response bias

  • liberal: more willing

  • conservative: less willing

• Discriminability

  • how far away is that procedure from chance performance line

To summarize, scientific efforts to reduce misidentifications hones in on which two issues? How were each of those issues resolved?

• The relationship between confidence and accuracy

  • Misidentifications were often made with high confidence in court

  • This is due to studies analyzing correlation but this was the wrong way to look at it

  • CAC analysis shows they are reliable when made with high confidence

• Simultaneous vs. Sequential lineups

  • Diagnosticity ratio showed a better outcome for sequential lineups

  • Chad Wixted used ROC analysis to show better outcomes for simultaneous lineups. ROC considers two things that diagnosticity ratio doesn’t:

    • discriminability

    • varying response bias (conservative and liberal)

• Moving to left/right on a single ROC refers to what concept?

• Moving to a higher/lower ROC refers to what concept?

• what is the name for the statistic that describes the relationship between hits and false alarms for a single point on the ROC?

• Response Bias

• Discriminability

• Diagnosticity ratio

what was the ebbinghaus curve / forgetting curve?

• Ebbinghaus tested his ability to retain information from a list of nonsense syllables and then plotted them on a graph. Over the course of a month, he was retaining less and less of it and it formed an L shaped curve on a graph. Interestingly enough, this same curve applies on different time intervals such as over the course of a day rather than a month

What experiment informed researchers of a potential permastore? What is the permastore theory of memory?

• Background:

  • Gave 733 people a spanish Test

  • Divided them based on how long ago they took Spanish and plotted it along a 50 year retention curve to see the relationship

• Data:

  • Graph shows a significant retention drop for groups 3 - 5 following the completion of spanish courses. This retention drop levels off.

• Conclusion:

  • Leveling off suggests some information must be entering a permastore.

How does the permastore theory conflict with the Ebbinghaus function?

• When put through statistics stuff to see if the permastore matches the Ebbinhaus function it shows a 91.4% similarity rating. The reason these two theories conflict is because the retention interval of the spanish study shows the exact same curve as a ebbinghaus curve. The Ebbinghaus curve applies no matter what time scale you are looking at. If you were to look at it over the course of a day, rate of forgetting would look the same yet in this instance it would be silly to suggest a retention interval over the course of a day suggests a permastore.

What are the two main explanation for why we forget?

• Decay (biological explanation)

  • memories consist of formation of synaptic connections, and forgetting consists of the natural loss of those connections

• Interference

  • new memories compete with old memories but do not damage their neural representation (psychological explanation)

    • retroactive interference

    • proactive interference

  • new memories disrupt the neural synaptic connections that represented old memories (biological explanation)

what is long-term potentiation?

• The synapse is physically changed following an experience and that change is the neural basis of memory.

  • change is the strengthening in the connections between two neurons

  • LTP readily occurs in hippocampal neurons, and memory formation is impaired if LTP is blocked

How is artificially-induced LTP researched? What is the outcome of that research?

• There is a recording of a post-synaptic neuron and there is a stimulation induced on the pre-synaptic neuron

• Results:

  • Neuronal response is generated from the stimulus in the form of a neuron firing —> Training of stimulus —> Larger neuronal response is generated from the stimulus which shows LTP (aka learning)

• What synaptic changes are associated with LTP? What term is attributed to these changes?

  • 1) How soon do these changes occur?

  • 2) Are they permanent?

• New receptors inserted into post-synaptic membrane

• Structural changes in post-synaptic neuron

  • more synapses

• The term is cellular consolidation

• 1) These physical changes in the synapse occur in a matter of hours after learning and are how neurons of the brain quickly code memory

• 2) The changes gradually disappear to some extent (decay) even in low-level organisms

What are the two types of interference?

• retroactive interference

  • forgetting caused by subsequent learning

  • new memories compete with old memories

• proactive interference

  • forgetting caused by prior memories

  • prior memories compete with new memories

Describe the research related to retroactive interference. What is the explanation for the results?

• Background

  • Experimental group learns word pairs A-B and A-C

  • Control learns A-B and unrelated C-D

  • Results show that control group exceeds experimental

• Explanation

  • cue-overload interference: retrieval cues lose effectiveness when spread out across multiple memories

  • This applies to above experiment because the experimental group had the retrieval cue A tied to both B and C while control group did not

Describe the research related to proactive interference.

• Background:

  • Example of this in class was three trials of naming 3 items of single category followed by counting backwards by 3

  • Over time performance decreases with increasing trial number

  • Can be ‘released’ from cognitive burden by being asked to memorize different category such as profession, flowers, or vegetables

    • interestingly enough professions is significantly easier than something somewhat semantically related to the original category like vegetables

• who were Muller and Pilzecker?

• what did they propose?

• how did they research this?

• what do the results of their research suggest?

• Background:

  • Proposed theory of consolidation

  • proposed the notion of retroactive interference (at the level of the memory trace or engram, NOT the cue)

• Experiment:

  • Memorize paired words followed by interfering list (either 17s later of 6 minutes later)

• Result:

  • Retroactive interference was GREATER when the interfering list was learned shortly after learning (17s) than when interfering list was learned after a long delay (6 min)

  • We find a temporal gradient to interference in the last study (more interference occurred if the second list occurred 17s after the first vs. 6 min.)

• Implications:

  • Shows evidence for CELLULAR consolidation

what does the temporally-graded nature of this study’s result imply about consildation?

• Consolidation is a continual process and as memories are more consolidated they are less likely to be forgotten

What is the difference between cellular and systems consolidation?

• Cellular Consolidation (small):

  • LTP in the hippocampus stabilizes (e.g., structural changes to the synapse)

• Systems Consolidation (big):

  • 1) Cortical Modules are wired strongly with the hippocampus

  • 2) Cortical Modules’s wiring with hippocampus weakens

  • 3) Cortical module disconnects from hippocampus and is memory trace is represented as the interconnectedness between cortical module

• what is the curious phenomenon of retrograde facilitation?

• what research supports this?

• Sleep, alcohol, and benzodiazepines inhibit encoding only

  • Encoding process selectively inhibited: Anterograde amnesia

  • Consolidation process intact, leading to: Retrograde Facilitation

• Sleep Study - Jenkins & Dallenbach

  • Two groups: Sleep and waking

  • information learned prior to sleeping showed an increased memory performance when compared to waking group

• Alcohol Study - Parker et al.

  • Subjects studied 30 slides of nature scenes

  • Following this, subjects consumed varying amounts of alcohol

  • Forced-choice recognition test given 7 hours later

    • Forced choice recognition: when presented with photos, you have to determine whether each picture was seen before (“old”) or whether it’s a picture you haven’t seen before (“new”). You must categorize the photo as either old or new.

  • Results:

    • People who consumed higher dosages of alcohol showed improved memory performance when compared to the group that didn’t

why/how does sleeping, consuming alcohol, and taking benzodiazepines all facilitate memory consolidation?