Psych 258 (Cognitive psychology).pdf

Dates not tested, pay attention to bolded concepts. Intro: How do you do science if you can’t directly observe? Many scientists began pursuing and/or constructing cognitive psychology, such as wundt, often considered the father of the field. They considered the mind a black box. The rise of be...

Dates not tested, pay attention to bolded concepts. Intro: How do you do science if you can’t directly observe? Many scientists began pursuing and/or constructing cognitive psychology, such as wundt, often considered the father of the field. They considered the mind a black box. The rise of behaviorism is due to the results being directly observable, thus being more “believable.” ○ Classical conditioning was a popular experiment in this field, consisting of Pavlov and his dog, and “little Albert” ○ Stick with what you can observe and “prove.” (Operant conditioning.) Rewards vs punishments. ○ Parsimony, simplicity is superior. This was the basis of behaviorism. Tubular rasa, blank chalkboard, no one has predispositions, thus one can easily predict how a person will behave, and “mold” them with operant conditioning. “All behavior is explainable.” Cogntives disagreed with tubular rasa. Maybe certain behavior should be studied more, as it (behavior) can maybe be influenced beyond simple conditioning. ○ For example, the rat maze experiment tackled tubular rasa. The rat went in the correct direction to get to the cheese, contrary to what the behaviorist thought of the rat going in the same direction it previously went. Latent behavior was evident from the rats creating a cognitive map. (Something is going on in the mind that isn't seen or displayed in overt behavior.) ○ In arguments over language acquisition, Skinner fully fronted the behaviorist view and stated that rewards alone were the reason for language learning. Chomsky disagreed with this, partially, he wanted to dive deeper into the mind. For example, children say things they’ve never heard before, therefore in-born nature is at play. The Paradigm shift, or major change in perspective, occurred during the cognitive revolution (shift away from behaviorism). The invention of the  computer also heavily influenced this, as cognitives thought of the mind as a computer, and how a better mind can build better computers. Information processing approach. Computers have storage memory which can be filled, does the human mind also have a limit to memory? (no, but it was argued.) ○ The dichotic listening task proved how the mind can only focus and listen to what it's attending. (They wanted to see if humans could absorb both audios in both ears equally, but only the attended, or “focused” ear audio was remembered, and the other one blocked out, kinda like a computer.) In the broadbent filter model of attention, the mind filters out specific input for further processing to memory, other input is blocked or “forgotten.” Process Models and Structural Models are the two basic forms of Models in Cognitive Psychology. ○ Process Models: Processes or steps involved that help to retrieve information or memory. (Input can either be delivered to short term, or long term.(Can be retrieved).) I remember the Capital of Canada (Semantic), but not where I initially learned this information (Life events). ○ Structural Models: (Remember the lobes and parts of the brain.) Structure of physical models such as the brain. Cognitive neuroscience: The nerve net appeared to be an interconnected network, each one connected to another. But, it was found out that they were not all connected, and had a tiny gap between each other. So if they're not connected, how do they communicate? Neurons are specialized cells to receive and transmit information through electrochemical means.  ○ Action potential of neurons. (All the same size.) One way to measure action potential is through microelectrodes, which measure individual action potentials. (Only used on non-human animals.) This measures the firing rate, how frequently do they occur? (Not the size.) Synapse, Neurotransmitters. (Hubel and Wisel 1960) Feature detectors are neurons that have been tuned (like a radio) in a way so that they are sensitive, or fire to a specific or preferred stimulus. (Eg. neurons that only fire when they hear a high frequency sound but not a low frequency sound.) ○ In their experiment, Blakemore and Cooper exposed certain groups of kittens to only vertical or horizontal lines, and when they were introduced to the other type of lines, their neurons would not fire. (Eg. People who grow up speaking tonal languages are much better at noticing small changes of vocal frequencies compared to English speakers.) Hierarchical processing is where the complexity of a visual object is dependent on how much it’s moved throughout the brain. ( One part of the brain might fire looking at a face, compared to random lines.) Mirror neurons fire when observing others do a task, and doing it yourself. (Letting someone move your body to properly swing a bat.) ○ It’s argued mirror neurons are important socially, and autistic individuals mirror neurons function differently, which is why they might not be affected by someone else’s visible emotions. Sensory Coding: All of them may be relevant in different situations.  ○ Specificity coding: The idea that the activity of a singular neuron codes for any given reaction in the brain towards a specific concept/target. Eg: If I see Sarah and a specific neuron fires, I can attribute that neuron to only and always react whenever it views/hears/etc Sarah. (One to one relationship.) We can only measure little bits of time, and not the entire brain. (I have evidence x is true, I don't have evidence y is true, this does not mean y is false.) ○ Population coding: Assumes there is a large number of neurons that fire towards a specific concept/target. Eg: 10 Neurons fire whenever they view/hear/etc Sarah. Action potential is all or nothing, but the firing rate can vary. So if the signal is loud, it means alot of neurons are firing. So the higher bracket does not mean the action potential is longer, but its firing more than the others. ○ Sparse coding: More than one neuron, but less than what is associated with population coding. Eg: 5 neurons fire when Sarah is present. Relative difference between 1 and a very large number. Localization of function: The idea that specific parts of the brain are responsible for specific function in the body. (Visual cortex is the specific area for vision, but also are some other parts of the brain. So if the visible cortex is damaged, they may not comprehend what they’re seeing, but still be conscious of their sight. Eg, blindsight. A person can see shapes, but not know that they're seeing them.) ○ Earliest indicator that localization of function was legit, was through the studying of individuals with brain damage. Broca’s Area (Language production, damaged means, you can understand what others say, but not produce comprehensible language.) and Wernicke’s Area (Primary auditory Cortex, comprehends language, so if damaged, affects your understanding of language, but can still speak.) Double association: If you damaged your thalamus, and found that that affected both vision and hearing, we wouldn't be dissociating that region from either of those processes. (Basically,  if the visual cortex is damaged and affected sight, we can infer the occipital cortex has nothing to do with vision, and vice versa.) Single cell recordings measure a single neuron reaction or firing to things. Fusiform face area (Taso et al) : when faces are shown, specific neurons fire in a specific region of the brain. When this region is damaged, people often have trouble identifying faces.(Prosopagnosia). Neural Stimulation: When jolts (fake action potentials) were delivered to a man’s FFA region, he was asked what was happening, all he stated everything was the same, besides faces becoming disoriented. ( Faces change shape like during psychedelics.) Brain Imaging: fMRI machines means the machine is functional, or using a different procedure, but is generally the same as MRI. When neurons become more active, they require more oxygen and glucose, which they get from increased blood flow to the area, this is than measured by the MRI. ○ Voxels are cubes of MRI images that are used in measuring specific brain activity. So while individual neuron activity cannot be measured, the general clumps of neurons can. ○ PPA responds to outdoor/indoor places (difficulty with navigation if damaged), EBA responds to pictures/parts of the body. In a study involving showing clips to participants having their brains scanned, they studied what parts of the brain became active depending on what is viewed during the clips. Scenes showing talking reliably activated the same parts of the brain in participants. In distributed representation certain objects automatically activate many parts of the brain, depending on past association as well. Eg, when viewing apples, the hunger, touch, vision, and taste part of the brain would be activated.  Even simple tasks require the activation of many neural regions to help aid in processing. (Seeing someone you like or dislike also activates different parts.) (If there's a conversation happening, and the drop of a water bottle covers up a segment of the person's speech, you can still infer the sound they made depending on their past speech patterns, or mouth sounds. Basically, clues are used to fill in blanks.) Distribution representation and localization of function are complementary to each other, regardless if it sounds contradictory. (The captain of the team is important, and if the captain is not good the team might be so bad, but the captain alone is not enough for the team to do good.) ○ No “one to one relationship”, broken region = cannot do the task anymore. PET scans are quite important in brain imaging. It also estimates neural activity with increased blood flow to a certain region like fmri, through fmri makes inferences how the magnetic properties of blood is changed, but PET is reliant on the injection of synthetic glucose traces. ○ PET scans are done less commonly these days, as it involves the injection of radioactive material. (You can also inject synthetic neurotransmitters like dopamine.) ○ Both pet and fmri have really good spatial resolution, but have poor temporal resolution. (Can't differentiate what's happening on a fine time scale.) (Not going to talk much about TMS.) ○ Transcranial Magnetic Stimulation inhibits the activation of neurons (Decreasing firing or vice versa.) It’s a newer invention, which is the mid field of altering a person's brain. (Pseudoscientists like this one, as they believe it can alter a person, like reversing depression. But little evidence is present of how good it holds up in the future.) If doing a study with a non-human animal, by damaging their brain, a more causal inference can be made. (If a task they could previously do, they no longer can after their brain is damaged, we  can infer that part of the brain has an effect on the task.) As with humans, correlation does not equal causation. Neural networks: Are formed through experience, memory, exposure, education etc. ○ Works similar to the internet. Even if you’re not using your internet, the connections are still present and working. ○ Depending on what you're doing moment to moment, the functional connectivity changes. (If you're doing homework, and your friend asks what you’re doing over the weekend, the neural connections switch to something more relative to your friends topic.) ○ Your brain never stops working, think of automatic functions such as your heart beating and lung breathing. Even when brain regions are at rest, or not in use, they're still active. The default mode network is when your mind starts wandering, either purposefully or unconsciously. (Being out of focus at the last minutes of class.) What is your brain doing when you’re doing nothing? Sometimes, mind wandering happens when we focus on things we perceive to be more important. (Evolutionary advantage?) As people age, they tend to mind wander less. Mostly because their brains become less active. (Linear decrease.) ○ Structural connectivity are physical connections. ○ Functional connectivity is the correlation between different functional parts of the brain which are related. ○ Resting-state functional connectivity is a particular task to see if two areas are connected. MRI machines are used to determine these analyses, by testing at the seed and test location. Than, the waveforms are graphed to see if their related or similar. Do we have to remember what parts of the brain does what Neural Networks: Certain areas of the brain are more active during specific tasks.  ○ Diffusion tensor imaging maps the major pathways in the brain. Also referred to as track-weighted imaging. This can help with finding how our individual brains react differently to different stimuli. (Musicians may feel stronger reactions to instruments and music.) Wiring is the process of viewing reactions branch to specific location, while fmri shows the end result. The Nature of Perception: Why are people presented with the same stimuli but can come to different conclusions? Experience dependent plasticity helps your brain have specific skills and fine tuning abilities. (Tonal language speakers and musicians.) When viewing a basic scene, people have knee jerk reactions and assume things immediately with reference to past experiences or memories, but that doesn't mean they’re correct. They don't have evidence that the dark spots on building A is a shadow, it could just be dark paint. (Think of optical illusions.) ○ Our sensory system feeds us data which we have to understand and conclude. Old women or young women? Rabbit or bird? Blue or White dress? We have to make sense of the puzzle. Perception is defined as the experience resulting from the stimulation of senses. ○ Image recognition technology is not as efficient as the human brain, but it can help in understanding how the human brain works. How did the computer get the picture wrong? The toaster was so bright and unnatural, the computer was distracted by it. Just like human perception, our visuals are distributed. Unfortunately, we don't have feedback telling us whether we were right or wrong. We live in a three dimensional world, but our retinas convert it into 2-dimensional. Inverse projection problem is the task of determining the object that created the image in your eye. Oftentimes things are covered up, so the brain fills it in, which varies from person to person. Viewpoint invariance, how one perspective makes it easier to identify an object. (Head front view of a fly is harder to identify, than a side view.) Bottom-up and Top down processing, is the act of how one perceives or comes to the conclusion of the stimuli they observe. With B-U, raw data that is transduced to higher levels of processing, is then analyzed in the brain. T-D relies more on memories, expectations, salience, etc. You're watching a magic show, and want to figure out the tricks. (So before it's even begun, you already have expectations on what you want to see. So you may pay more attention to little things. ○ In a lecture, you may be going back and forth with B-U and T-D. Direct perception theories are reliant on B-U, as the stimuli is analyzed in the brain, whereas Constructive perception theories focus on T-D, with analyzing being based on experience, expectations, etc. ○ You'll have different experience getting a vaccine depending on your view or expectations on it. ○ When analyzing blurred images, people may come to a conclusion on what they perceive. (The blurred product is actually the same in all pictures.) But due to their connection to what's happening in the image, we may perceive it as something else. (The blob on a person’s feet might be shoes. The blob on the sidewalk may be a pedestrian.) P cool huh? ○ Audiovisual speech perception is how visuals may affect our perception of the audio. (Ba-Ba or Fa-Fa depending on mouth movement, McGurk Effect.) Imagine your chewing on something that's supposed to be crunchy, but you can’t hear the crunch, so you perceive it differently. ○ Speech segmentation, when listening to a foreign language, you may not know where the words end, as it just sounds like one long broken word, so SS is where speech is divided. Helmholtz theory of unconscious inference, states that our perceptions are based on our personal history of experience. We don't realize we're doing this. The likelihood principle is related where it assumes the only possibility it imagines due to regualarity is the only right one. (The red paper is not cut behind the blue paper.)  Structuralism broke down complex ideas into simpler concepts, thus stated perception is built from simpler concepts. Gestalt rejected this, and stated that perception was based on the laws of perceptual organization. ○ Gestalt would say, if the dots were arranged differently, we would see a dog. Animation is just still images put together. Where is the motion? (Technically, there is no motion,) Phi phenomenon. The goal is more than the sum of its parts. ○ The principle of good constitution bias’s our perception, to make us assume the branches follow a smooth plan. Law of pragnanz is the principle of simplicity. The olympic symbol looks like 5 overlapped rings, but it could also be 9 shapes. ( Due to us not seeing those shapes regularly, we won't assume there are 9 shapes.) ○ Principle of proximity and similarity makes us group close and similar objects together, respectively. They look like one coherent meaningful representation. It assumed that these principles helped us in our survival in the past, (Works similar to a mental shortcut.) Physical regularities occur naturally in nature, so our perception is affected by it. Horizontal lines are more common than oblique lines in nature, which explains why we can perceive horizontal lines a bit better, as we’re “tuned” to them. (Oblique Effect.) ○ Think of the kitten test with horizontal and vertical lines. Except the oblique effect is more milder. ○ Our system is tuned to respond to things we encounter more frequently in nature. ○ Light from above representation is when our brain sees a normal and upside down image similarly. (Concave and Convex). Our natural light source is from above (the sun),so upside down light is still perceived to be “normal.” (Convex or concave are viewed the same.) ○ In Semantic Regularities is where you expect to find something in a certain environment. (You expect to find a professor and students in a classroom. Probabilistic likely to see in a kitchen, such as sink and a fridge. Scene Schemas)  We can recognize a pan on a stove much better than a printer on a stove. ○ Bayesian Inference, Likelihood of something that is going to happen (Expectations) vs the actually stimuli in front of you. Prior: Maybe you do something that is usually a mistake, (you google what your symptoms might mean, and come to a scary conclusion that your cough could be a cold or heartburn). Likelihood: Then you talk to your doctor who says a cough could mean a cold or lung disease. Conclusion: If your previous experiences are compatible with new information or likelihood, you're more likely to believe in it. (The cough means a cold.) When comparing the four approaches, the first three (Bayesian, Helmholtz and Regularities) approaches rely on previous experiences shaping your present view, which is susceptible to dynamic change. (Top-down.) But, the Gestalt approach is believed to be more innate and not driven by experience. Though we can't really know how much its based on predisposition or experience. ○ Picture (b) does not look like W and M, but picture (a) does. Your brain recognizes (a) as letters because of past experiences and coherence. Our brain gets better at recognizing things we come across more frequently. A newer computer is faster than an older one. Same thing with children as they have a lot of neurons compared to adults and can learn languages more easily. (Plasticity.) Perceptual expertise is when a person has more experience in a certain field and can recognize things much more efficiently. ( An art expert can pick a fake painting from an authentic one better than the lay person.) ○ A scientist trained participants to recognize “Greeble.” Where the ffa began recognizing it after becoming tuned to it through training. Initially there was no response. (FFA is naturally trained as we come across faces our entire lives. People who have face-blindness do not show ffa activation.) Behaviour: There's a difference in watching stimuli through a screen vs real life. When you engage in action, look around, touch things, it's much different than being still and static. (When observing things from different angles, you receive more useful data and interpretations.) It connects with the larger idea of how perception truly works. A Lot of our perceptual system is fine-tuned to make specific movements. ( I pick up a supposed empty can, but end up nearly dropping it because it's full. Your brain automatically adjusts your strength and grip to certain objects based on past experiences.) In Lesioning, you damage a part of a non-human animal's brain after training them in a specific task to see if that part of the brain has any effect on the specific action. ○ In the Object discrimination problem, and landscape discrimination problem, the monkeys who had their ventral streams and dorsal streams compromised could only do the other task and not their own. (Double dissociation.) Midterm Cutoff is on friday, last lecture before midterm. Perception and Action ○ DF injured her temporal lobe. She could insert the card into the slot, but not match the card in the same way of the slot from a distance. ( As the location, navigation (motor cortex) was still present) ○ When you are doing certain things that require/activate part of the brain in an “automatic” way, you don't realise you gained the skill in the first place. (Reaction to stimuli is automatic.) If you're watching baseball players swing their bat, neurons in your motor cortex will fire if you choose to swing a baseball bat yourself one day. Don't want to ask favour from someone you offended, but want to ask from someone you “think” is in a good mood. Mirror neurons at play, while autistic individuals can't figure out the emotional states of others. Attention You are prioritising your limited resources in a way you think is strategic. (Is it possible to process everything happening in the environment? Advantages?disadvantages?) Disconnective to certain spatial locations.  Selective: Attending to one thing (top-down way)(Maybe something salient catches your attention), while ignoring others. (Should you focus on your book, a barking dog, or the angry phone call? Depends.) Can be intentional or unintentional. Divided attention: Paying attention to multiple things. (Multi-tasking. Read an email while watching a lecture?) Sustained attention: How easy is it to focus on things presented to you? (Easy to focus in the beginning 10 mins, compared to 50 mins.) Same task loses attention over time, not always intentional. Spatial attention: Attention over space, Dichotic Listening task: Listen to one ear, and ignore the other ear. ( Two different audios playing in both ears.) Ability to filter out irrelevant stimuli. Also called the cocktail party effect, because if there's a lot of conversations occurring, you have to focus on one and ignore bits and pieces of others. ○ People back then did not believe that attention could work this way. In Broadbent's filter model, participants could not recall the information in the unattended ear. (Only attended information will transform into memory.) Also can be considered an “early selection model”. ○ There's a lot of stimuli in the environment which your brain recognizes. But it gets filtered out to only the things you pay attention to/detect, and thus, retained in memory. In Gray and Wedderburns experiments, participants generally combined information they received from both ears. (Dear 7 Jane.) When asked to report what they heard in the left ear, they started incorporating things from the other ear as well. Combining words to make a coherent idea represented by language. Dear Aunt Jane makes more sense to them than Dear 7 Jane. (Broadbent's filter model could not explain this.) ○ Moray also found similar findings as information leaked through the “filter.” So the filter is not as perfect as thought. Triesman’s Attenuation model: The attenuator analyses messages in terms of their strength, and divides into stronger and weaker paths. Information from the  attended ear seeps through at a much stronger strength than the unattended ear. (Reaches unconscious awareness. Also depending on salience.) Different words will have intrinsic meaning. If you're hearing a sound of speech/words related to you (like your name), it would be easily detected even in weak form. Both of these models can be considered early selection models. But certain people argued attention came later on. Do you have any more classes today? You would think about it, and then respond. One can either believe attention comes early on, or much later. (Believed to occur relatively early on.) Late selection models: Mckay argued that attention occurred after meaning was already established. One ear heard an ambiguous sentence, “They were throwing stones at the bank.” The other ear heard a biassing word, river or money. Later on they were tested on the meaning of the ambiguous sentence. (Alot of reasons someone might be predisposed to think of or the other. Money banks are mentioned much more in cities.) Found a systemic relationship between the outcomes of the answer they gave, and the biassing word they heard. Attention, Automaticity, and Competition: Load theory of attention: You're on a long train ride, and have nothing to do. Due to your boredom, you unintentionally eavesdrop. Even though you might not want to. When you're on your phone, your focus is only on it and not your surroundings. In high-load tasks, such as playing a video game, your brain can't process that and a podcast in the background effectively. In low-load tasks, or doing nothing, your brain looks for things to process. ○ A full sponge cant pick up more water. Your attention is like an empty glass, filling up to a point you cant fill it in more. In the easy option of the task, there are minimal distractions and complexities, and you have to look for a letter. (N,O,O,O). In the hard condition, you have to find the letter in a group of different letters. (K,M,P,I,S). Easy trails have a quicker response time. ○ In the second part, a distracting cartoon character is also placed in the picture. Cartoon characters slows you down more in the easy condition, than in the hard condition. A bit confusing, but think about it. If you're  doing a hard task, you're using more resources on that specific task, and don't have any remaining residual resources to analyse the cartoon character. (Locked in.) Stroop task: Ink colour and shape colour dont give a person any complication. (I can analyse the ink colour and shape colour separately with ease.) If it’s ink colour and word meaning, which is a colour. It’s not hard to read the word's meaning and ignore the ink colour. It's difficult to ignore the word's meaning and only pay attention to the ink colour. Why? Incompatible representation with the stimuli you should be paying attention to. Colour and shapes are totally separate. But names of colours and actual colour are related. (Ink colour is Blue but word meaning is Yellow. You read yellow, and start thinking of the ink colour yellow as well.) Different people have different success in inhibiting irrelevant stimuli. Some people are better at blocking out background noise than others. ○ People who speak multiple languages have a smaller interference effect. I see something black, and can either automatically think of Noir or Black. ○ A target can either show up in the left or right side of a screen, and you must press the correct direction button (compatible). Or, you can press the opposite direction button (Incompatible). It’s easier and faster to press the compatible direction button. Incompatible stimulus dimensions are conflicting and confusing. Tracking attention: In terms of eye-tracking, if you look at something, your attention travels with your vision (overt). But of course, you can pay attention to something happening in the corner of your eye without moving your head (covert). Saccades, Fixations, Fixation durations, Regressions, Pupil dilation. ○ You’re reading a page on a book, but realise a bit later on, you didn't comprehend what you just read. So you go back and reread it. Eye-tracking can be used to track people's attention on certain things, and can be used in developing products that can quickly “grab their attention.” ○ Red dot is more easily found when surrounded by green dots, than when surrounded by multi-colour dots. (Bottom-up?) ○ If an alarm was ringing in a hallway, it would be more salient in the quiet than on top of other ringing alarms.  ○ Red diamond captures attention due to its high salience, even when not intending to look for it. In an experiment people were looking for stop signs, they were good at finding it when the stop sign was located in a typical/regular area like an intersection. When they show up in unexpected locations, like in the middle of the road, they miss the stop sign. ○ Prioritising how they distribute their attention in line with their expectations. (Top-down) ○ Can you process everything in your environment? We probably can't comprehend everything in the environment. Load theory and the finite resource of attention. If you live in the desert, you may end up noticing more unexpected stop signs compared to the average person, but miss more of the typical locations of stop signs. (Also top-down, plus trade off.) ○ Back to scene schemas, people fixate on out of context objects for longer, like a printer in a kitchen, which was found with eye-tracking. Take a longer time to process and make sense of. When looking for lost car keys, you're more likely to look on places where keys are commonly found like a table, instead of on top of a painting or under the rug. Back to eye tracking, people’s eye tracking is aligned with their intended goal. (When asked to guess the characters ages, people more likely looked at faces.) ○ Think about how much data you can get if playing for a team and receiving eye-tracking data. Certain plays are paying more attention to opposing players, goalie could be paying attention only to the puck, etc. Doesn't tell you exactly why they're doing it, but eye movement and how they distribute their attention can be inferred. In an experiment, participants are fixated on a cross on the middle of the screen, and are told to click quickly which location the target is located (left or right.) (Testing covert attention as you're only supposed to be looking at the cross and not move your eye.) ○ There will be either a valid or invalid trial, in which the valid is where the arrow is pointing, and vice versa for the invalid trial. Spatial attention, is attention fixed to locations, or objects that happen to occupy the location, or both? Depends.  ○ In a study, the goal was to prove how attention is bound to objects, rather than the location. A cue was located on one side of the rectangle. Same-object advantage, if any point on a coherent unified object is cued, or more relevant, your attentions spreads throughout the boundaries. In a study of covert attention, participants were asked to covertly shift their attention to different areas. ○ Need enough data to make predictable models. Huge pool of data showcasing shopping behaviour, we can look for average or generalised patterns. Useful in marketing. When doing initially difficult tasks and becoming highly practised at them, you start doing them automatically, or automatization. ○ A lot of us can type on a keyboard without knowing where the keys are. But initially, it was difficult. ○ Imagine you are shown a bunch of random letters, and asked to find a target that changes randomly. Sometimes the target is present, sometimes it’s not. People start at 50/50 chance level, but after a lot of practise, they become good at it. Is using a cellphone while driving dangerous? Automatization is when you’re driving and when you arrive, you don't remember the drive. Phones affect this greatly, making you more susceptible to crashing due to not noticing unexpected outcomes on the drive. Like a cat running in front of your car.