Keep up to date with everything happening at LDPL

April 2017

LDPL Tech is a revolutionary online learning platform that uses artificial intelligence and data science to improve education for the teacher and the student. LDPL Tech’s groundbreaking platform recommends learning topics for each individual student in order to address any gaps in knowledge or skills and teachers have access to shared content and real-time data insights into their students’ learning. Your Life, led by a board of directors chaired by Edwina Dunn, co-founder of dunnhumby and CEO of Starcount, is a three-year STEM campaign to ensure the UK has the Maths and Physics skills it needs to succeed in today’s competitive global economy. The Your Life campaign engages young people by creating inspiring video content and by running activities such as memorable visits to the U.K.’s most exciting STEM workplaces.

So, how are LDPL Tech and Your Life joining forces?

On LDPL, students access learning content and artificially intelligent algorithms plot the most effective route through the material. LDPL provides courses mapped to the National Curriculum and Your Life create videos that are used as learning material throughout these courses. The videos provided by Your Life have been embedded into captivating mini-courses for extracurricular and lifelong learning and are also dotted throughout LDPL’s GCSE Maths and Physics courses. LDPL Tech and Your Life are continuing to work together to create exciting courses to inspire students to study STEM subjects by providing engaging material and a personalised route through their learning.

April 2017

LDPL Tech, an artificially intelligent online learning platform, has been selected for participation in Pitch@Palace. Vote now to help them win the People’s Choice Award!

Pitch@Palace was established by The Duke of York to support entrepreneurs in the development of their business ideas. Entrepreneurs present their innovative ideas to an audience of CEOs, investors, angels and influencers. Attendees have the ability to catapult the entrepreneurs’ ideas to the next level, whether they be potential mentors, investors, or business partners. The theme for Pitch@Palace 7.0 is ‘human tech’, so all of the entrepreneurs’ business ideas explore the potential impact of technology on improving our lives.

LDPL Tech’s purpose is to improve education for the teacher and student. This is achieved by using artificially intelligent technology to learn how each student learns and provide them with a personalised learner path based on their unique learning needs. By learning on the platform, students generate data which is then presented, in real-time, to the educator and provides insight into their strengths, weaknesses, completion and effort level. The platform also auto-marks short, formative assessments and selects suitable learning content for the student, helping to reduce the admin-heavy workload faced by teachers.

People can vote for their favourite business idea by voting for the People’s Choice Award, which is now open. The winner will be announced at St James’s Palace on 25th April. Vote for LDPL Tech today at http://pitchatpalace.com/vote/.

Priya Lakhani, Founder CEO of LDPL Tech said:

“Our aim at LDPL is to improve education for everyone. We really hope people support us by voting for LDPL in the People’s Choice Award so we can spread our message and help even more students and teachers!”

The Duke of York said:

“I am immensely proud of the achievements of the Entrepreneurs in the Pitch@Palace 7.0 programme. In less than three years, there have been over 500 pitches at events all over the United Kingdom and they have shone a light on the diversity and imagination across the country, clearly demonstrating that pursuing an idea or dream can be realised with knowledge and determination. I wish all those taking part in the People’s Choice Award and the final of Pitch 7.0, every success. ”

Feb 2017

Neuromyths in education are nothing new. The Guardian has highlighted four of the most common, which our Cognitive Neuroscientist discusses.

There is no surprise that teachers should be interested in psychology and neuroscience. As a cognitive neuroscientist, it is encouraging to see so many teachers trying to incorporate evidence from the science of the brain into their lessons. However, neuroscience is anything but simple and the prevalence of so-called ‘neuromyths’ in the classroom is cause for concern.

The Guardian (2016), TES (2016, 2014, 2013), The New Scientist (2014), the BBC (Radio 4 programme, 2013) have all featured articles on this problem, calling out the common neuromyths and dispelling them. And yet, they persist.

Do we really need to worry? So what if people think that you can be left-brained or right-brained? How bad can it be to believe in learning styles even though there is no evidence to back this up?

Well, expert Paul Howard-Jones says it can be pretty bad, actually. He claims that belief in these neuromyths can hinder effective teaching. Usha Goswami, a researcher at Cambridge University (PDF), suggests that the best way to teach new material is through a range of styles. This contradicts the neuromyth that we have a specific learning style (Visual, Auditory or Kinesthetic). If children only receive learning materials in one of these styles — due to the mistaken belief that this is beneficial — their learning has been impeded.

So, what should the teacher interested in neuroscience in education do?

Luckily enough educational neuroscience is a rapidly growing area of research and it’s beginning to produce some interesting results. In 2014, the Wellcome Trust partnered with the Education Endowment Foundation (EEF) to fund further research into promising educational strategies. There is plenty to interest teachers looking to incorporate cognitive neuroscience understanding into the classroom. Currently being investigated are projects into Growth Mindset, Working Memory, Spaced Learning and Gamificationin the classroom.

They have also compiled a Teaching and Learning Toolkit which is a summary of educational research on teaching 5–16 year olds and is constantly updated to reflect the latest understanding of teaching strategies.

LDPL is committed to incorporating the best teaching and learning research into our platform. Our recommendation engine currently includes adaptivity based on active retrieval practice (spaced learning and the testing effect) to encourage robust memory formation. We include cognitive messaging around our site to encourage growth mindset, resilience and grit to develop independent, confident learners. We are also piloting an investigation into the role of emotions in learning and long-term memory.

Cognitive neuroscience can provide useful strategies for improving learning outcomes. But we have to be aware of neuromyths, of overselling the evidence and of drawing conclusions that aren’t warranted. If we can do this, then we can drive powerful change in the classroom and beyond.

Feb 2017

Growth mindset is a term coined by Professor Carol Dweck at Stanford University. She distinguishes between growth mindset and fixed mindset. A growth mindset of intelligence is the belief that intelligence can change over time: it is possible to increase your abilities by applying effective learning strategies. In contrast, a fixed mindset is the belief that intelligence is a fixed, innate attribute that cannot be changed: the ability you have at the outset is as good as it will get.

We see mindsets at work everyday, both in the classroom and out. Typical fixed mindset statements look like this:

“I can’t do maths”

“I’m just not creative”

“Oh, I’ve never been sporty”

In all these statements there is a fixed mindset declaring that there is no control over ability.

It’s easiest to see growth mindset in action around games. If you lose a level during a videogame, you typically start it again trying to do better this time:

“Oh, I nearly had it that time! This time I’ll get it”

When you start a game, you believe that you are capable of winning, even if you don’t win straight away. In other words, you have a growth mindset towards the game.

Why does it matter?

Let’s pretend we have a class of two: Alex and Sam. They are given a problem to solve and they both get it wrong. What happens next?

Well, Alex has a fixed mindset so he believes that his ability to solve this problem is already set. There is no point in trying to solve the problem again, no point in learning more about it, trying to understand other ways to solve it. He is either intelligent enough to solve it the first time around or he isn’t. He didn’t, so there is nothing more he can do.

Sam has a growth mindset which means she believes that with effectively applied effort she can solve the problem. She will keep on trying: trying new strategies, trying to understand more; trying to solve it. Sam believes that she will be able to solve the problem eventually.

So, why does mindset matter? Because it alters what we do when we encounter set back. Ultimately, mindset matters because it has a strong relationship with outcome. Students who keep trying are more likely to achieve better outcomes.

What’s the evidence?

Carol Dweck has done comprehensive research into the relationship between mindset and learning outcomes. In 2013, she conducted a review showing that mindset interventions result in improved learning outcomes for children who have a fixed mindset (Yeager, Paunesku, Walton & Dweck (2013)). Interventions which change the type of praise a student receives have been shown to encourage a growth mindset. Additionally, interventions which teach children how the brain learns or which focus on the try-fail-try-again routine of famously successful figures are effective.

How does LDPL incorporate this research?

LDPL encourages a growth mindset in a couple of different ways.

Students get sent personalised cognitive messages which encourage resilience and growth mindset. These messages offer learners effective learning strategies based on their current performance and effort levels. They also inform the students about how the brain learns to more implicitly encourage a growth mindset. All messaging around results is also grounded in growth mindset research to encourage a mindset which can improve learning outcomes.

Additionally, LDPL has a course which teaches students how the brain learns. Dweck’s research, among others, showed that understanding of the basic functions of memory and learning can help learners see that abilities are not innate. This course was developed in conjunction with HRH Duke of York, as part of iDEA. It is aimed at learners aged 11–14, but there is something there for everyone. Why don’t you give it a go today and see what you can learn?

Citations

Blackwell, L., Trzesniewski, K. & Dweck, C., 2007, ‘Implicit Theories of Intelligence Predict Achievement Across an Adolescent Transition: A Longitudinal Study and an Intervention’, Child Development 78(1): 246–263

Yeager, D., Paunesku, D., Walton, G. & Dweck, C., 2013, ‘How Can We Instill Productive Mindsets at Scale? A Review of the Evidence and an Initial R&D Agenda’, White Paper for the White House meeting, Excellence in Education: The Importance of Academic Mindsets.

Dec 2016

With 1.3 million children underperforming in the UK (Ofsted, 2016) and 74% of teachers considering leaving the profession due to unmanageable workloads (TES, 2016), it is clear that the current education system is facing some serious challenges. LDPL has been created by teachers, students, parents, software developers and neuroscientists with the purpose of improving education for all.

LDPL’s revolutionary technology has taken centre stage at the TechCrunch Disrupt Battlefield 2016 to showcase its platform which leverages artificial intelligence and big data technology. LDPL Tech’s technology learns how each individual learns, adapting their learner journey to reflect their learning needs. LDPL uses a range of adaptive variables including pace of learning, difficulty levels, modality preference and effectiveness, spaced learning algorithms and item response theory.

Priya Lakhani OBE, LDPL Tech’s Founder CEO says, “At LDPL, we are passionate about improving education. Our platform uses advanced technology that makes a real difference to both the teacher and student. I am thrilled that this has been recognised by TechCrunch Disrupt Battlefield!”

So, how does it work?

Students access learning material through LDPL. LDPL hosts a multimedia library of content including, GCSE maths and English language, and maths and English Functional Skills, Entry Level 3, Level 1 and Level 2, all of which is mapped to the curriculum. Teachers can also easily add their own subjects and content.

Artificially intelligent technology then learns how each student learns, providing them with a personalised and adaptive learning journey, constant, formative assessments and instant feedback. All messages students receive are tailored to their experience and are grounded in cognitive neuroscience, designed to encourage a growth mindset and resilience.

Teachers and SLT are presented with real-time actionable data that supports evidence based teaching and reduces time spent on planning. By tracking homework, auto marking and finding resources, LDPL reduces the admin burden faced by teachers. The deep insights presented to educators show their students achievement, knowledge, skills and performance against assessment objectives, identifying their strengths and indicating where interventions may be necessary.

LDPL Tech is currently being used by more than 10,000 students, with several more secondary schools and colleges receiving their logins in January. LDPL Tech is also a finalist for the BETT Awards and Learning Awards and Founder CEO, Priya Lakhani, recently won the Special Achievement Award at the Mayor’s Fund Awards.

LDPL will be exhibiting at BETT 2017 as part of BETT Futures, stand F60. During the exhibition, LDPL will be giving live demos of the platform as well as short seminars on artificial intelligence, cognitive neuroscience and data in the classroom, among other topics. In addition, LDPL will be offering schools and colleges that sign up during BETT, free access to the platform for the rest of the academic year plus a significant discount for the following academic year.

To find out more, please email info@LDPL.tech

 Nov 2016

What are good data practices? What should be avoided?

Liz Macfie, Data Scientist at LDPL, gives us some insight.

Over my years in data science (and also those as a mechanical engineer before that) I’ve had to learn good data practices, often through making mistakes. I’m going to share some of the more easily avoided slip-ups I’ve seen/done with the hope that this might especially help organisations without a dedicated “data” team (or at least without someone as outspoken as me!).

Track everything. Immediately.

Resources can be spread thinly when starting a new project, especially if there are tricky deadlines to hit. Regardless, data gathering has to be an immediate priority, even if nothing is done with it straight away. We can guarantee that in 3 months time some bright spark will ask how current user behaviour compares to past user behaviour.

I’d also recommend having a kick-off meeting with all personnel who might eventually want to use the data. Perhaps developers don’t know the whole story and would have left out the tracking of a metric that later became a key business priority.

Avoid vanity metrics

It’s understandable, especially when just starting a new project, to be fascinated with raw user numbers. We want to track every action they take, we want to know how many there are, we want to see live activity. This is perfectly fine and can create a shared excitement as screens go up showing what users are doing in real time. However, we have to go deeper than these metrics for business decisions.

A very simplistic example: Let’s say the most important part of our product is a button, and every time a user presses it, we magically get some money. Obviously we want to measure engagement with this button, so we create a graph showing the number of button presses each day.

We spot that the button press numbers go down at the weekend and start fretting over what this could mean: do users only want to pay on weekdays? Someone then thinks to plot number of daily button presses per number of daily users and gets this:

It turns out there is no problem with button engagement, there are just fewer users at the weekend. Reporting the “vanity metric” (number of button presses) rather than the actionable metric (number of button presses per user) was unhelpful.

Keep numbers accurate

We’ve all been there. The quarterly report is due, and we’re the tiniest of fractions below a particular target. Knowing that there are three kinds of lies (lies, damned lies and statistics) we work out a way to “massage” the data so that it falls on the correct side of this arbitrary line. Ethics aside, there are a couple of major problems with this:

1. Anyone else wanting to produce the same numbers has to know about our statistical manipulations otherwise there will always be discrepancies, and trust me — if there’s one thing any board hates, it’s discrepancies.
2. If we actually improve the next quarter, but still don’t hit that target, what do we report? The accurate apparently lower number, or do we engage in more data trickery to also bring this second number above the target, so correctly reporting an improvement?

I say, the more honest you are with data in all reports, the more grateful future-you will be.

Verify all results

I don’t think I’ve ever regretted taking a little longer to check numbers I’m about to report, but I’ve certainly often regretted moving too fast and reporting an inaccuracy. This can so frequently be avoided by having multiple ways to generate the same statistics: perhaps we send website data to two sources; perhaps we store the same information in databases in two slightly different ways; perhaps we carry out a calculation again with the steps in a different order.

In addition to this, there should always be an idea of whether variations being reported are “significant”, but that’s a topic for another post.

And finally… my personal pet peeve…

Throw out the pie charts

Just no. Can we please stop with these now? They are a tool to teach students about circle sectors or to show the proportion of uneaten pizza… they are not a valid data visualisation!

All of these are very basic non-technical ways to start a healthy data-driven culture within any company. After setting these principles, the fun begins!

 Nov 2016

LDPL’s Cognitive Neuroscientist, Alice Little, gives us an insight into what cognitive neuroscience really is.

Let’s start with the ‘cognitive’ bit: what is cognition? Cognition refers to all the stuff that goes on in your brain when you think. It refers to all the stuff that happens when you process the world via your senses. Acquiring knowledge, learning things, perceiving. Any mental action or process is a form of cognition.

Cognitive science is the multidisciplinary field that studies cognition. Understanding the memory process, or perception, decision making, problem solving, language acquisition or emotion regulation are all the domain of cognitive scientists, amongst many other things. If it’s a mental process that involves thought, then cognitive scientists are all over it.

Now for the neuroscience bit. Neuroscience is the study of the physical structure and function of the brain (and nervous system). It is the study of neurons, of the chemicals in the brain, of the electricity flows in the brain. Neuroscientists might look at specific instances of brain damage to deduce what that area of the brain is involved in; they might use imaging techniques; they might look at the function of an individual neuron or a highly specific neurotransmitter. They might investigate human brains, primate brains, rodent brains or even the more primitive brain-like structures in simple organisms. But the thing that brings all neuroscientists together is that they are studying the brain itself.

So, what do you get when you cross a cognitive scientist with a neuroscientist? Well, you get a specific method for studying cognition. Cognitive neuroscientists investigate any aspect of cognition with direct reference to what is going on in the brain: we study the brain to understand what the mind is doing.

In short, cognitive neuroscience is the study of brain processes to understand how the mind works.

And how is LDPL using Cognitive Neuroscience?

LDPL’s general principle behind incorporating cognitive neuroscience is the same with any features we design and implement. We trust our data to tell us what is effective and what isn’t effective. We will design implementations of various theories and then let the data tell us whether they are successful for learning. If not, we will iterate and try again. Ultimately, we are attached to no cognitive dogmas; we have no vested interest in seeing one theory succeed or fail. The only thing we are driven by when considering the inclusion of cognitive theories into LDPL is this:

Does it make the learning better?

There is nothing revolutionary about applying cognitive theories to pedagogical practice, but that doesn’t mean we can’t be innovative with how we do that. In the coming blogs I will give a more detailed picture of a couple of the cognitive principles we are currently using in our software and what the evidence is for their benefit for learning, so look out for What’s the big deal with Growth Mindset, anyway? and Let’s test out the Testing Effect if you’re interested in learning more.

 Oct 2016

Nadya Thorman is the Chief Operations Officer at LDPL and previously taught with Teach First for three years. Here she explains how technology can be used to benefit teachers.

When I started my Teach First journey in 2010, I, like many participants, had absolutely no idea what I was getting myself into. I had applied to Teach First after being shocked by the appalling social mobility statistics I had come across in my university studies. Teach First offered an opportunity to address the issue; I could make a real difference to the lives of young people, and I could do it immediately.

Within a few weeks of the school term, that starry-eyed, naïve, young graduate was exhausted and dispirited; I was beginning to doubt whether I could convince Abdul to bring a pen to class, let alone transform the educational opportunities of the students I taught.

Teaching was, and still remains, the most difficult thing I have ever done. Don’t get me wrong — I loved teaching; I loved the students who made me laugh every day, and I loved the colleagues who worked tirelessly to improve the life chances of others. I was even pretty good at it, at least so others told me. But it was hard, and it didn’t feel sustainable, so, after three years and much soul-searching, I made the decision to leave the profession.

Unfortunately, I was not alone in finding teaching a challenge: in a recent survey of teachers, 82% of respondents reported that their workload was unmanageable. I have returned again and again to this dilemma: if good teaching is beneficial to society (and I think we all agree that it is), then how can we make it a more sustainable career? So when I was asked if I would join education technology startup LDPL Tech, which aims to improve learning outcomes while reducing teacher workload, I jumped at the opportunity.

Teachers know how to make learning happen: we know that students benefit from immediate and constructive feedback; that differentiated materials enable students, who learn in different ways and at different speeds, to make similar progress; that accurate data can be used to identify key strengths or weaknesses more quickly; that for some students, a ‘fixed’ mindset holds them back, and that more than teaching them new skills, we need them to understand that they have the ability to grow their intellect. We also know that doing all these things for all our students is a monumental task. Our intention at LDPL is to make this task easier.

LDPL’s learning platform provides students with a personalised learner path at the same time as arming teachers with learner data, so that they are the very best educators they can be. The platform uses artificial intelligence technology, cognitive neuroscience and big data insights to begin to understand how students learn best by analysing students’ behaviour on the platform, e.g. time active, accuracy of answers, media studied, response time, etc. According to the DfE, data management and marking are the biggest drivers of ‘unnecessary and unproductive tasks’ in a teacher’s day. So LDPL aims to reduce teachers’ workload by automating the marking of students’ work and presenting learners’ data in an easy-to-use dashboard. Teachers are provided with a real-time view of how their students are progressing so that they can intervene as and when necessary.

We have big dreams for the future of education and are busy working with innovative schools and colleges in order to achieve them. We are always looking to collaborate with educators who are excited about the possibilities that technology can bring to education. If you fit the bill, or are interested in finding out more, please email info@LDPL.tec.

We can’t wait to work together to make a difference to the world of education.

May 2017 

Are robots going to steal your job? That’s unlikely says Kat Arney

March 2017 

March 2017

Classroom artificial intelligence programme gains global recognition

Feb 2017

Learning by keystrokes

Feb 2017

AI, big data and the future of your classroom (page 29)